Incentive Regulation and Distribution Network Performance: A Dutch Case Study (2000–2024)

Incentive Regulation and Distribution Network Performance: A Dutch Case Study (2000–2024) | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Incentive Regulation and Distribution Network Performance: A Dutch Case Study (2000–2024) Paul H L Nillesen, Michael G Pollitt This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7601011/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 26 Apr, 2026 Read the published version in Journal of Regulatory Economics → Version 1 posted 4 You are reading this latest preprint version Abstract This paper assesses the long-term effects of incentive regulation on the financial and operational performance of Dutch electricity distribution system operators (DSOs) from 2000 to 2024. Using a proxy for regulated income—Constructed Regulated Revenue—we evaluate the outcome of yardstick-based benchmarking across seven regulatory periods. Constructed Regulated Revenue per connection rose from €284 in 2000 to €381 in 2024, a real-term increase of 1.2 percent per year. This growth, however, masks substantial heterogeneity: from 2000 to 2017, productivity improved by 1.9 percent per year, but between 2018 to 2024 this dropped to -3.6 percent per year. The backward-looking design of the regulatory framework limited incentives for anticipatory investments. Expenditures rose sharply as DSOs responded to grid congestion and the demands of the energy transition, eroding earlier cost reductions. We conclude that while incentive regulation in the Netherlands was effective in reducing costs during the early periods, it did not adequately support long-term investment under evolving system conditions. The results underline the importance of forward-looking regulatory mechanisms that align cost efficiency and pass-through cost with future grid needs. JEL Codes: L94, D24, Q48, L51 Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1. Introduction The Netherlands adopted a price control review approach to the economic regulation of its electricity Distribution System Operators (DSOs) in 2000. The Dutch DSOs have been subject to regulatory overview and yardstick competition for 25 years, based on an ex-ante price cap with an exogenous efficient cost level. Yardstick competition ensures that during the regulatory period all efficient costs at sector-level are recovered through tariffs. The regulatory objectives of the national energy regulator (ACM) over the full period have remained in place: (i) provide network operators an incentive to operate in an efficient manner, (ii) prevent network operators from charging tariffs above the (efficient) cost level, (iii) allow network operators to earn an appropriate return on invested capital, and (iv) encourage optimal quality of transport. There have been a number of studies examining the impact of regulation on costs, quality and investments in The Netherlands. For example, Nillesen & Pollitt ( 2007 ) analyse the impact on consumer welfare from the failed first distribution price control review. Baarsma et. al. ( 2007 ) discuss the economic and legal implications of ownership unbundling of distribution networks from supply companies. Niesten ( 2010 ) makes regulatory recommendations to facilitate the integration of distributed generation. Nieuwenburg et. al. ( 2020 ) assess the financial performance of network operators between 2012 and 2019. Montfort et. al. (2024) analyse the impact of regional differences related to the energy transition on yardstick regulation outcomes between 2012 and 2020. This paper takes a longer-term perspective by presenting a detailed empirical assessment over a 25-year period. It examines the question how incentive regulation – particularly yardstick benchmarking – has affected the financial, operational, and investment performance of electricity DSOs in the Netherlands from 2000 to 2024. The analysis is timely from both a policy and regulatory perspective given the focus on electrification and the concurrent investments that are required to meet increasing demand because of the energy transition. The focus of the past 25 years has been on lowering end-user tariffs through efficiency incentives. As more distributed renewable energy has been connected to the network and more demand is being electrified (such as the electrification of transport), network investments have been increasing substantially in the past few years and are projected to rise significantly in the next decades (see Netbeheer Nederland, 2024 ). The rapid electrification in the Netherlands has also meant that investments have lagged demand, which has resulted in congestion and significant grid constraints (see for example, Financial Times, 2025 ). Further, the ACM has announced that it is considering changing the regulatory framework from 2027 onwards, from the current output-based yardstick competition system to a more forward-looking input-based system, to account for increasing investments and the need for more forward-looking guidance (ACM, 2023 ). This would bring Dutch regulation more in line with the forward-looking practices implemented for energy network price controls in Great Britain by Ofgem (see Duma et al., 2024 ). The structure of the paper is as follows. Section 2 outlines the regulatory context. Section 3 gives an overview of the data and the revenue proxy we construct. Section 4 provides the results from the analysis. Section 5 explores the limitations of backward-looking regulation and draws policy lessons. 2. Incentive Regulation in The Netherlands Incentive-based regulation is commonly referred to as RPI-X (or CPI-X) regulation outside the US and was first suggested by Littlechild ( 1983 ). 1 This form of regulation allows average tariffs (0r revenues) to increase inline with inflation less an efficiency factor (the X-Factor), which reflects the potential for relative productivity improvements. The linking of regulated prices to performance is known as Performance-based Ratemaking (PBR) in the US. 2 Under both PBR and RPI-X regulated firm retain extra profits for the duration of the regulatory period. This form of price-cap regulation thus decouples profits from costs by setting maximum average tariffs (or revenues), by creating a regulatory lag (see Vogelsang, 2002 and Joskow, 2014 ). To set the appropriate X-Factor or performance metric, many regulators have applied benchmarking techniques to compare the regulated firm’s actual performance against some pre-defined reference or benchmark performance. As highlighted by Schleifer ( 1985 ) this approach has the theoretical advantage that it reduces the informational asymmetry that occurs in rate of return (ROR) regulation by reducing the regulator’s reliance on the firm’s own costs. The most disputatious part of price-cap regulation is the benchmarking technique used to determine the productivity gap and how outcomes are translated into X-Factors. Regulators have adopted a variety of benchmarking methods to arrive at X-Factors. There are broadly three different approaches: (i) historically observed costs (US ROR regulation), (ii) projections about firm-specific efficiency (European and Australian regulators), and (iii) industry-wide (or wider economy) productivity indices. European regulators have generally adopted firm-specific projection-based benchmarking methods to calculate X-factors, such as Data Envelopment Analysis using international data from other regulated entities. PUCs that adopted PBR have tended to use either historically observed costs (ROR) or industry wide productivity measures such as Total Factor Productivity (TFP) to calculate the efficiency requirements (see Jamasb and Pollitt, 2001 ). One of the main advantages of this approach is that it does not involve a direct comparison of the regulated firm with another firm’s costs. This can expose the regulator to legal challenges based on unfairness given incompatibility (as documented in Nillesen and Pollitt, 2007 ). The Netherlands adopted a price control review approach to economic regulation in 1998 with the Electricity Act. 3 Originally the first price control period was to be from 1 January 2001 to 31 December 2003. The first consultation document for this price control review period was published in July 1999 (DTe, 1999 ). The outcome of the first regulatory period was disappointing. In terms of the length of overrun in the process (the first X-Factor was set in summer of 2000, but final X-Factors were only agreed in May 2003) and in the unmet initial expectations, it compares unfavourably to other price reviews that regularly take place in for example the UK, Norway, Australia, and Chile. Of the initially promised savings of €511mln only €384mln turned out to be accurate. The subsequent court case defeats (for the regulator) and previous allowance of the “ no reformatio in peius” meant that only €209mln of the €384mln was achieved. The ACM subsequently abandoned individual benchmarking and introduced weighted-average yardstick competition based on Schleifer ( 1985 ) in the second regulatory period (from 2004), where network operators are compared to mimic competition on cost levels (CPB, 2000 ). This approach is therefore less about finding theoretical “efficient costs”, but more about creating incentives to lower costs in a repeated game. The approach chosen by ACM attempts to capture the output generated by the DSO and when combined, the sector. The DSO supplies a portfolio of products and services, such as connection services, transportation services, and feed-in services, each with an associated tariff. The multiplication of the quantities of these products and services with the respective tariffs gives a “composite output” for each DSO. 4 The aggregate output, for the whole sector, is then used to calculate average sector costs. This is simply defined as the total costs of all DSOs (based on a TOTEX approach) divided by the total aggregate output of all DSOs. In principle all operational costs are included in the yardstick including network losses (technical and administrative), except for transmission costs (TenneT) and transmission costs from other DSOs. Depreciation and the allowed return (WACC) are based on the Regulated Asset Base (RAB). The sum of operational costs, depreciation (based on the RAB), and the allowed return (WACC * RAB), is the TOTEX for the sector. At the start of each reporting period, the RAB is updated to account for depreciation and investments. The sector-weighted average cost per unit of output forms the basis for the yardstick against which the performance of individual DSOs is assessed and used to determine the regulated revenue and tariffs. Dynamic efficiency is introduced by extrapolating historic productivity growth. As such, the combined costs of the sector are recouped through tariffs within a regulatory period, although an individual DSO can earn a surplus or have a deficit if their individual unit costs are lower or higher respectively. For certain cost components (e.g. transmission costs or network losses) the ACM has announced that it will allow recovery earlier in the regulatory period. For example, transmission costs from TenneT are difficult to forecast. In the past these costs were determined ex-ante with a two-year delay for determining actual costs, leading to significant pre-financing by the DSOs (without any compensation). Other costs, such as investments related to the energy transition and the feed-in of decentralized generation will be determined ex-post. In the past there have been numerous legal challenges to price control reviews. Some of these challenges have resulted in adjustments to the methodology or the interpretation of the exact application of the methodology. This means that in certain years or regulatory periods there is the possibility of under or over recovery, and subsequent adjustments in the next price control review. 3. Methodology and Data The Netherlands currently has six electricity DSOs. Three of which (Liander, Enexis, and Stedin) account for more than 98 percent of all connections. At the start of regulation in 2000 there were nine electricity DSOs. Three of the smaller DSOs were acquired by the big three players during the period we analyse. At the start of regulation, in 2000, the big three DSOs had a combined market share of just under 94 percent. The big three DSOs themselves are the result of multiple mergers between smaller utility companies. 5 Although separately benchmarked initially, they were subsequently viewed as single players by the regulator as operations, investment, and management were consolidated. 6 To assess the financial and operational performance of the sector over the last 25 years we have collected data from various sources. Given the timespan under consideration there is no single data source. We have therefore created a dataset using multiple sources, such as company annual reports and company publications, regulatory filings and regulatory publications, and publicly available information. The data is based on the three largest DSOs – Liander, Enexis, and Stedin. Table 1 provides a description of the variables. [ Table 1 – description of variables sources ] We construct a proxy for the sector-level regulated revenue – the Constructed Regulated Revenue – by applying the building block approach. This includes operational costs, transmission costs (TSO cost pass-through), network losses, depreciation, and the regulated return on invested capital. See formula 1. [ Formula 1 – Constructed Regulated Revenue ] $$\:{Constructed\:Regulated\:Revenue}_{i,t}={Operating\:Costs}_{i,t}+\:{Transmission\:Costs}_{i,t}+\:{Network\:Losses}_{i,t}+\:{Depreciation}_{i,t}+\:{Return}_{i,t}$$ Where i is the DSO and t the year. The Return is calculated using formula 2. [ Formula 2 – Return ] $$\:{Return}_{i,t}={Real\:pre\text{-}tax\:WACC}_{t}*{Fixed\:Assets}_{i,t}$$ Where i is the DSO and t the year. Where possible we use data directly from annual reports or regulatory filings. Where data is not available or missing, we interpolate or make estimates based on reference figures. For example, in some years transmission costs and network losses are not reported separately, but as one category. Another challenge is the fact that there is limited data available for the DSOs prior to the compulsory ownership unbundling, when data was aggregated at group levels. [ Table 2 – Collected data 2000–2024 ] Table 2 provides an overview of the collected data. This includes the financial variables related to the Constructed Regulated Revenue, such as operational costs, fixed assets, and network losses. We also present operational statistics, such as number of connections, network length, volumes transported, and SAIDI. Table 2 also reports data related to the regulatory framework, such as the pre-tax real WACC, CPI, the regulatory period, and the connection-weighted average X-factor per regulatory period. Finally, we present related data for the analysis, such as wholesale electricity prices, Total Factor Productivity for the Netherlands, electricity demand, installed capacity of solar PV and onshore wind, and the number and estimated capacity of EV charging stations. The Constructed Regulated Revenue is compared with the actual or estimated reported revenue. Over the full 25 years the total difference between our estimate of regulated revenue and the reported revenue is negligible (less than 0.5 percent). However, in certain years there are differences of more than 5 percent. Thus, in some years we are over- or underestimating the actual regulated revenue. This is due to several factors. In some years the data is incomplete or missing (leading to assumed values) or the level of reporting is different (e.g. at group level versus at DSO level). There are timing effects we do not consider. For example, certain costs, such as transmission charges are based on regulatory estimates with actual costs recouped in later years. In addition, there have been numerous legal proceedings that have resulted in adjustments and ex-post corrections to correct for over or under recovery across regulatory periods. Our estimate for the allowed return uses the reported asset value in the annual report (based on IFRS 7 ) as a proxy for the RAB. The regulatory accounting rules deviate in certain aspects from IFRS. In addition, at the start of the first regulatory period the RAB was treated as one asset class with one remaining depreciation period. 8 Over time the RAB value therefore deviates from the underlying IFRS-based valuation of the invested capital. The main deviations seem to occur during periods where the projected development of costs by the regulator deviate from the actual development, or when structural changes occur. An example of this is in 2020/21 at the end of the seventh regulatory period, when the projections for 2022–2026 were made. This did not include the severe impact on wholesale energy prices from the Russian invasion of Ukraine. Wholesale electricity prices were more than three times higher in 2021 than in 2020, and more than doubled again in 2022. This had a notable effect on the value of network losses, which were significantly higher than estimated in 2020/21. This leads to a significant difference between our Constructed Regulated Revenue and the reported revenue based on tariffs set by the regulator. For example, our Constructed Regulated Revenue for 2022-24 is approximately 6 percent higher. In general, these deltas between actual and projected revenue are corrected and recouped in subsequent regulatory periods. The other reason for significant deviations is structural in nature. 9 Liander and Enexis were fully unbundled at ownership level from NUON and Essent respectively in 2009. Stedin was only fully unbundled from Eneco in 2017. Although we have attempted to correct for this based on publicly-available information, it is likely that the actual costs deviate in magnitude and timing. 10 In addition, in 2008 the 110/150kV networks owned by the DSOs were transferred to TenneT, the TSO. This is also one of the reasons why the third regulatory period was shortened to one year instead of three years. The financial data we present and use for our analysis therefore provides a directional view of the impact of incentive-based yardstick regulation on the financial performance of the sector as a whole and smoothes out the ex-post calculation effects. The Constructed Regulated Revenue is therefore a proxy of the average connection costs borne by customers connected to the distribution grid. It represents the “average network bill” for end-users (as represented by connections). In this case the end-user is not just a typical household but includes connections to small and medium enterprises and separately-metered EV charge points for example. Given the fact that there is no single data source and our dataset comprises of multiple sources, with some estimations and interpolations when data is missing, we only examine the development of the financial and operational performance at sector-level and do not compare the performance of Liander, Enexis, and Stedin with each other. 4. Results Average network costs over time To compare the results over time, we deflate the financial data to 2000 prices using the Consumer Price Index (CPI). Table 3 summarises the main results from our analysis and Fig. 1 shows the development of the Constructed Regulated Revenue per connection from 2000 to 2024. [ Table 3 – Summary of main results ] [ Figure 1 - Constructed Regulated Revenue per Connection, 2000–2024 (real 2000) ] The Constructed Regulated Revenue per connection rose from €284 in 2000 to €381 in 2024, a real-term increase of 1.2 percent per year. This growth, however, masks substantial heterogeneity: from 2000 to 2022, average costs declined by 0.8 percent annually, but surged post-2021 due to two primary drivers. First, the energy crisis following Russia’s invasion of Ukraine drastically increased wholesale prices, affecting network losses (see for example Tertre, 2023). Second, TSO transmission costs tripled between 2021 and 2024 due to elevated investment levels. In 2022 the average wholesale electricity price in the Netherlands was eight times higher than the price in 2020. This commodity cost flows into the costs for end-users through the cost for compensating network losses (which are approximately 4.5 percent of transported volumes at the distribution level 11 ) at both the DSO level and via transmission charges from the TSO. Losses per connection were twice as high as average in 2023 for example. Second, transmission costs – which are passed-through – have increased due to increasing investment levels by the TSO. Transmission costs per connection in 2024 are almost three times higher than in 2021. In the period up to 2022 average connection costs decreased by 0.8 percent per year or 14.9 percent. Excluding the pass-through costs from the TSO gives a view on the core DSO-related costs. Over the full period the Constructed Regulated Revenue per connection decreased by 0.4 percent per year. The average distribution network connection costs were approximately 10 percent less in 2024 than in 2020 (in real terms). [ Figure 2 - Constructed Regulated Revenue components, indexed (2000 = 100) ] Operational costs per connection declined by 21.6 percent between 2000 and 2018. Since 2018 operational costs have however been increasing. In 2023 operational costs per connection were the same as at the start of regulation, and in 2024 14.3 percent higher than in 2000. Operational costs are strongly related to the size of the asset base, the level of investment, and the share of distributed generation in the network. We discuss the impact of increasing investment levels later. The most striking change over the full period is the drop in the allowed return per connection – almost halving from €83.4 per connection in 2000 to €42.8 in 2024. This significant decrease is the result of a combined effect of a lower allowed rate of return by the regulator and higher capital utilization. The regulatory real pre-tax WACC dropped from 6.7 percent in 2000 to 3.5 percent in 2024, as interest rates declined over this period (see Fig. 3). [ Figure 3 - Average return per Connection and the Real pre-tax WACC ] Investments and the energy transition At the same time capital intensity increased between 2000 and 2017, with fixed assets per connection dropping by 25 percent. This improvement in capital efficiency – adding connections at a higher rate than expanding the capital base – further put downward pressure on the allowed return per connection. From 2018 onwards average fixed assets per connection have increased by more than 31 percent, with the average fixed assets per connection in 2024 equal to the level at the start of regulation in 2000. Overall, the impact on the total allowed return of the increase in the asset base is outweighed by the significant drop in the allowed return (+€155mln versus -/-€349mln in allowed return, respectively). [ Figure 4 – Change in fixed assets (real) ] Figure 4 shows the change in fixed assets over time. Following a period up to 2008 where depreciation exceeded investment levels and the asset base declined, there was a period between 2009 and 2016 when the asset base expanded by circa €190mn per year. From 2017 investment levels have been increasing year-on-year. Between 2023 and 2024 alone, the three DSOs invested €2.3bn. The Capex-to-Opex ratio also shows how investment levels are increasing. At the start of regulation, the ratio was less than 0.5 but has been close to or greater than 1 since 2019. Projections by the association of network operators (Netbeheer Nederland) show that investment levels will increase further in the next decades because of increased electrification and higher renewable penetration. 12 [ Figure 5 - Fixed assets, feedin capacity (solar and wind onshore) and off-take capacity (charge point and heat pump) over time ] Figure 5 shows the growth in renewable feed-in capacity (solar PV and onshore wind) and the growth in energy transition off-take capacity (charge point infrastructure and heat pumps) compared with the asset base of the three large DSOs in the Netherlands. Wind onshore installed capacity is currently 7GW (12 percent CAGR 2000-24). Solar PV has been even more successful, with an installed capacity of 30GW. 13 The installed capacity of solar PV in 2024 (23.4GW) was almost twenty times higher than a decade earlier in 2014 (1GW). The growth in distributed generation accelerated in 2017. The annual growth rate between 2000 and 2016 was 13.6 percent, whereas between 2017 and 2024 this almost doubled to 25.6 percent per year. The growth in EV charge points shows a similar pattern, following the sharp growth in the number of EVs. 14 The annual growth rate from 2015 – when the first reliable data is available – is more than 33 percent. There are currently almost 1mn charge points in the Netherlands. This includes home charging, public charging, semi-public charging, office charging, and high-powered charging. We present data that excludes home and office charge points, given that these connections do not require grid reinforcements and are “behind-the-meter”. Using average capacities for each category, we estimate the total installed capacity of these charge points. This has grown from 337MW in 2015 to 3.5GW in 2024. A further example of the electrification of energy demand that impacts electricity network operators is the switch from natural gas to heat pumps for heating. There has been significant growth in the number of installed heat pumps (air and water-based). From 49 thousand units in 2015, there are now 2.3mn units installed – an annual growth rate of more than 50 percent. This is still less than half compared to Europe’s leading country Norway with 632 per 1000 households. 15 The installed capacity of these heat pumps has grown from 354MW in 2015 to more than 13GW in 2024. The energy transition is responsible for the sharp increase in investment levels by the DSOs. It is not just direct and indirect investments in the network infrastructure, such as new lines, transformer stations, and grid enhancement, it also increasingly requires investment in supporting technologies and operational capabilities to deal with more active networks (e.g. bi-directional flows) and more volatility. Alliander has established a separate System Operations unit – like in transmission – given the increased active nature of their grid. The regulator has also recognized the impact of the energy transition and is considering moving to a more forward-looking input-based system to account for increasing investments related to the energy transition (ACM 2023 ). There have been several studies examining the impact of the energy transition on network costs. 16 In a study by Montfoort et al. (2024) using data from all the Dutch DSOs from 2012–2020, the authors find that energy transition variables, such as solar PV, wind onshore, and EV charging points, have a significant and substantial impact on unit costs. 17 Wangsness & Halse ( 2021 ) using a panel of more than 100 Norwegian DSOs between 2008–2017 estimate the effect of growth in EVs on distribution network costs. They find that increasing penetration of EVs increases network costs. Just & Wetzel ( 2020 ) find that distributed generation is a significant cost driver for German DSOs. Their results indicate that a 10 percent increase in distributed generation capacity leads to a total cost increase of about 1.6 to 2.3 percent – although not necessarily on a per unit basis. There are also technical and theoretical studies into the impact of EVs. Gupta et al. ( 2021 ), find that an EV penetration rate of 33 percent results in significant upgrades to transformer stations and line reinforcements. Fernandez et al. ( 2010 ) estimate that integration of EVs requires additional investments in distribution networks up to 20 percent. Gust et al. ( 2024 ) apply a theoretical model for designing distribution networks with significant demand coincidence (e.g. from EVs and heat pumps) to a large sample of Swiss DSOs. They find that high demand coincidence (e.g. by charging at the same time) increases average network cost by 84 percent (159 percent in the worst-case scenario where the coincidence factor is 1). Additionally, the impact of demand coincidence is greater on large networks and networks with connection density. However, analysis by Lundgren & Vesterberg ( 2024 ) of 179 Swedish DSOs between 2014 and 2021 shows that increasing small-scale generation and an increasing number of EVs do not significantly affect the technical efficiency of DSOs. Although there has been strong growth in the installed capacity of distributed generation and EV charge points, Dutch electricity demand has remained relatively flat. Our data shows that the average transported volume per connection has remained relatively constant between 2015 and 2024 (circa 10MWh per connection). We also find that network length per connection has remained constant (circa 35m per connection), suggesting that the grid has been expanding in capacity rather than utilisation. Overall Dutch electricity demand increased between 2000 and 2008, but has been relatively flat until the COVID-19 crisis. Dividing national demand for electricity by the number of connections, shows that demand has been declining since 2010 (from 15MWh in 2010 to 12.5MWh in 2024). Part of this reduction in demand is the result of energy efficiency, deindustrialisation, and the shifting of production – via solar PV – to behind-the-meter. The quality of the distribution network has remained stable, even with cost pressure from lower tariffs, increased investments, and the increase in intermittency. The average SAIDI has fluctuated between 20 and 30 minutes 18 , with an average of 22.2 minutes between 2004 and 2024. This SAIDI is extremely low when compared internationally. In Europe the average reported SAIDI is around 174 minutes, based on a selected group of DSOs for each capital city. 19 Productivity growth over time To assess productivity growth, we compare the change in output (number of connections) with the change in input (operational costs and capital costs, excluding transmission costs and network losses). This is a gross output or value-add approach, rather than the more classic Total Factor Productivity (TFP). Ideally, we would like to examine the utilisation of the network over time, and not just the number of connections, given that load factors and capacity utilisation also drive costs. [ Figure 6 - Productivity growth indexed (2000 = 100) ] The average productivity growth of the sector was 0.3 percent per year between 2000 and 2024. Over this period annual average operational cost productivity growth was − 0.8 percent and annual average capital cost productivity growth was 1.2 percent. This is the result of an increase in operational costs and increase in the allowed return driven by higher investment levels, whilst the growth in the number of connections remains constant (1.1 percent per year). The overall increase in productivity is in large part the result of the decrease in the allowed rate of return (see Fig. 2 and Fig. 3). This WACC-effect allows the capital productivity to increase. As discussed earlier, from 2018 the investment levels of the sector started to increase (see Fig. 4) because of the energy transition. Comparing the period before (2000–2017) and after the acceleration of the energy transition (2018–2024), shows a significant difference in productivity growth. Overall average productivity growth of the sector was 1.9 percent (2000–2017) compared to -3.6 percent (2018–2024). Operational cost productivity growth per year was 1.2 percent between 2000-17 but fell to -5.5 percent between 2018-24. As for the capital cost productivity, this dropped from 2.3 percent per year before the energy transition acceleration in 2018 to -1.4 percent between 2018-24. The sector compares favourably with the TFP of eight sectors across 11 European countries (see ACM 2020 ) over the full period, which finds an average productivity growth of 0.5 percent between 1995 and 2017. The developments up to the energy transition inflection point are also positive compared to other studies. Just & Wetzel ( 2020 ) analyse the cost efficiency of German DSOs between 2011 and 2017, using a dataset with 450 DSOs. They find an average cost reduction potential in the range of 12–18 percent taking both transient and persistent productivity factors into account, which is what we find for the same period on a per connection basis. Amundsveen & Kvile ( 2017 ) find average annual productivity increase of 1.3 percent between 2004 and 2014 for Norwegian DSOs – across three regulatory periods. In our sample we observe a similar increase (1.9 percent). They find evidence of a positive impact on efficiency from “smarter” networks. In addition, larger DSOs have higher average productivity – possibly related to scale economies with deploying smart technologies in networks. Ajayi et al. ( 2022 ) analyse the productivity growth of distribution networks in Great Britain and how changes in incentive mechanisms have influenced productivity. Over the 1990/91–2018/19 period the find productivity growth of approximately 1 percent per year. Ramos-Real et al. ( 2009 ) using a panel of 18 distribution companies in Brazil between 1998 and 2005 find annual productivity growth of 1.3 percent. The productivity of the Australian electricity distribution sector, on the other hand, decreased over 2006-15 period at an average yearly rate of 1.4 percent (AER, 2024 ). Following this, it trended up over the period 2015–23 with 0.9 percent per year. The improvement is the result of lower operating costs. Impact on costs and performance by regulatory period Although the Constructed Regulated Revenue is not perfectly comparable on a year-by-year basis given missing data and necessary estimations, the underlying changes in reporting, and the timing effects of regulation, we nevertheless analyse the impact of the regulatory periods on the financial performance of the sector. The results should be treated with caution and provide directional information, rather than an accurate impact assessment. Table 4 shows the changes in the different revenue components between the regulatory periods. [ Table 4 – Change in underlying components of the Constructed Regulated Revenue per Regulatory Period ] As discussed earlier, Table 4 shows the impact of the lower WACC on the composition of the revenue per connection. Whereas in the first regulatory period, the return component made up 31 percent of the average cost per connection, in 2024 that has dropped to 18 percent. The share of operational costs has conversely increased in importance, from 38 percent in 2000 to 48 percent in 2024. The depreciation component has been stable over all the periods. Losses remained relatively stable, except the sixth regulatory period when DSOs benefited from low wholesale electricity prices. The average network loss compensation in the current regulatory period are similar to at the start of regulation. In the third and sixth regulatory periods there is the biggest downward correction in allowed return in absolute terms, circa €15 and €25 per connection respectively. In the seventh regulatory period operational costs per connection show a big increase – €10 per connection – following the increase in investments. Figure 7 presents the indexed development of the DSO sector revenue using the X-factors set by the ACM and compare this with the development of our Constructed Regulated Revenue. 20 This gives a view on how accurately projected revenues, based on X-factors, track actual costs, and whether there is over or under compensation. [ Figure 7 - Constructed Regulated Revenue vs. ACM Projected Revenue (index 2000 = 100) ] In the first regulatory period the sector reduced costs quicker than the X-factor target, generating a cost-efficiency gain of €105mln relative to the projected revenue. Sector costs stabilise in the second regulatory period, whilst allowed revenues decrease due to the 2.7 percent X-factor. The shortfall in revenue in the second period equals €296mln. In the third regulatory period there is a limited correction, with a negative X-factor of 0.7 percent. 21 This results in a gain of €62mln for the sector relative to the projected revenue (2000 figures). The sum of the first three regulatory periods is a total revenue shortfall of €130mln. That is, sector costs were higher than allowed revenues. This shortfall is equal to approximately 1 percent of total sector costs over the three regulatory periods. Given some of the data issues we noted earlier, it seems plausible to assume that revenues matched costs across the first three regulatory periods (2001–2007). In the fourth regulatory period (2008–2010), the X-factor increases significantly to 5.2 percent. However, sector costs rise during this period, leading to €530mln shortfall between the allowed revenues and the actual underlying sector costs. To correct for this, the X-factor in the fifth regulatory period becomes negative (essentially allowing tariffs to increase). This leads to a surplus of €393mln for the sector. The reset of the starting cost base in the sixth regulatory period provides a higher starting point, but with a positive X-factor of 4.4 percent, allowed revenues decline again. The combined effect of these three regulatory periods is a €60mln shortfall – equivalent to 0.3 percent of total sector costs over the three regulatory periods. Based on this outcome, we also assume that allowed revenues matched sector costs between 2008–2016. The start of the seventh regulatory period in 2017 coincides with the inflection point in the underlying investment trend due to the energy transition. It also marks the first five-year regulatory period – in theory allowing for a more long-term and stable tariff income outlook. In addition, the real pre-tax WACC is further lowered from 4.3 percent at the start of the period to 2.8 percent at the end of the period. Although allowed revenues decline, sectoral costs do not. This results in a shortfall of approximately €1bn (~ 10 percent of total sector costs over the 5-year period). The X-factor for the eighth, and current, period takes this into account and is negative (-3.2 percent). However, the accelerating investment agenda, higher operational costs, and increased loss expenses, increase – rather than decrease – the discrepancy between underlying costs and allowed revenues. Although the regulatory period ends in 2026, the current shortfall is already more than €1.5bn – circa 20 percent of the total costs for the current three years of the regulatory period. Table 5 summarises the findings per regulatory period and compares this with some of the underlying drivers of the change in costs. [ Table 5 – Change in Constructed Regulated Revenue per Connection per Regulatory Period ] Average installed renewable capacity triples from the sixth to the seventh period and from the seventh to the currently still ongoing eight regulatory period. EV charging capacity shows a similar pattern. As for heat pumps, this is still at the start of the s-curve. As previously discussed, investment levels also start to increase. Whereas in the sixth regulatory period the fixed asset base grew by €406mln, in the seventh period it increased by €3.6bn, and in the current regulatory period it has so far grown further by €5.1bn, with two years remaining. In a backward-looking regulatory regime, it is difficult to observe underlying changes in a timely manner – and it requires judgement whether an observed change is transient or structural, and in the case of the investments in the energy transition in the Netherlands, whether they are accelerating. Yardstick competition has been well documented in prior literature when sectoral costs are stable, and changes occur gradually, and the magnitude is limited. It is less effective in situations where costs and revenues are more volatile, large structural shifts occur in short time frames, and when the pace also differs between DSOs (ACM 2012 ). There has been a lively discussion whether the regulator should have altered its approach and whether there was sufficient evidence that investments would be rising significantly, and that ex-ante yardstick competition was not providing incentives for DSOs to invest ahead of demand (Hensgens et al. 2021 , Nieuwsuur, 2022 ). Benefits of yardstick regulation To assess the overall benefit of yardstick regulation, we compare the Constructed Regulated Revenue with a counterfactual revenue in absence of regulation. We assume that network losses, fixed assets, and depreciation are the same in the counterfactual situation. Operational costs per connection are assumed equal to 2000 and scaled with the number of connections in each year. 22 For the allowed return we follow McKinsey (2002) and assume that the cost of equity remains stable over time. 23 We use the average of the low and high estimate of the cost of equity estimates from the ACM for the first and second regulatory period as the basis for the full period (6.0-9.1 percent). We then apply the estimated cost of debt by the ACM using a 60 percent leverage ratio. We then discount the difference between our Constructed Regulated Revenue and the counterfactual using the prescribed societal cost-benefit discount rate to assess government policy changes (2.25 percent real) (see PBL, 2022 ). Based on our analysis we estimate the total benefits, in Net Present Value (NPV) terms, from yardstick regulation for the period 2000 to 2024 to be equal to €2.6bn or equal to €294 per connection (both in real terms 2000). Interpreted as a perpetual annuity, the estimated NPV of €294 corresponds to approximately 2.5 percent of the Constructed Regulated Revenue per connection at the start of regulation. This implies that, on average, the regulatory effect over the period 2000–2024 is equivalent to a permanent efficiency gain of about 2.5 percent of yearly revenues per connection. Our findings are significantly lower than what other studies have found. Haffner & Meulmeester ( 2005 ) estimate the total cumulative benefit of regulation at €1.1bn for the 2001–2006 period. This is calculated by estimating the value of the tariff reductions without considering the counterfactual or taking the net present value. They also find that in real terms the energy bill has decreased by 12 percent over this period. In our analysis we find a decrease of 7 percent. Mulder & Plug ( 2009 ) calculate that in 2010 the cumulative benefits of regulation were equivalent to €6bn. Here also there is no clear counterfactual, no discounting, and it seems that the authors assume that current tariffs would remain in place in absence of regulation. 5. Discussion and Policy Implications The evolution of the Dutch electricity distribution sector over the past quarter-century offers a compelling case study of how economic regulation, technological shifts, and policy transitions interact to shape network performance, costs, and productivity. This analysis, spanning 2000 to 2024, reveals a sector that initially achieved substantial efficiency gains and productivity growth under a robust regulatory framework but now faces growing strain from external shocks, accelerated investment needs, and regulatory lag. The story is not one of linear progress, but of a sector and regulator adapting—at times reactively—to a rapidly transforming energy landscape. At the start of regulation in 2000, the constructed regulated revenue per connection stood at €284 (in 2000 prices). Over 24 years, this increased to €381, reflecting a modest 1.2 percent annual real growth. However, this aggregate figure conceals substantial internal variation. From 2000 to 2022, network costs declined by nearly 15 percent in real terms, due in large part to declining cost of capital, and growing capital efficiency. It is only from 2021 onward that this trend reversed. This reversal is the result of the underlying acceleration in the energy transition, with significant growth in renewable capacity from solar PV and wind onshore and significant growth in electrification of transport and heat, which started around 2017, leading to a significant year-on-year increase in capital investments. The investments by the DSOs were then amplified by investments in the transmission network (transmission costs tripled in three years). And although these costs are passed-through, they impact end-user bills. The energy crisis in 2022 further (temporarily) pushed up costs, such as compensation for network losses. This dichotomy over the full period is also clear when looking at the productivity of the sector. Overall productivity growth was 0.3 percent per year. However, between 2000–2017 it was 1.9 percent and between 2018–2024 it dropped to -3.6 percent. One of the most notable drivers of lower revenues per connection has been the consistent decline in the allowed return on capital. The regulatory real pre-tax WACC fell from 6.7 percent to 3.5 percent, mirroring broader macroeconomic trends in interest rates. This translated into a 49 percent reduction in the allowed return per connection—from €83 to €43. Early in the regulatory period, capital intensity was also decreasing, fixed assets per connection fell by 25 percent from 2000 to 2017, reflecting improved capital efficiency. These developments combined to significantly reduce capital costs, particularly in the first three regulatory periods. From 2018 onward, however, this picture began to shift as investments started to increase and drove an upswing in capital intensity. Fixed assets per connection increased by 31 percent between 2018 and 2024, returning to 2000 levels. Operational costs followed a similar but lagged trajectory. They declined steadily by over 21 percent between 2000 and 2018 but have risen since. By 2024, operational costs per connection were 14.3 percent above their 2000 level, reversing two decades of efficiency gains. This increase is partly structural, linked to the operational complexity of integrating variable renewables, managing bi-directional power flows, and accommodating demand-side technologies such as heat pumps and EV chargers. DSOs are no longer simply maintaining a passive infrastructure—they are operating active, intelligent networks. The regulatory framework—anchored in yardstick competition and the X-factor approach—has generally delivered cost discipline and incentives for efficiency. However, its backward-looking design has proven inadequate during periods of sudden structural change. The analysis of regulatory periods shows that while surpluses and shortfalls largely balanced out in the early phases, the last two regulatory periods saw substantial under-compensation: a €1 billion shortfall in the seventh period (2022–2026) and a further €1.5 billion shortfall to date in the eighth period. These shortfalls are equivalent to 10–20 percent of sector costs in the respective periods (in previous periods the shortfall was less than 1 percent). They arise from misalignment between allowed revenues and actual cost developments, particularly in the face of surging investment needs and volatile pass-through expenses. This misalignment is not merely a financial concern for DSOs—it has real implications for the energy transition. If DSOs are under-incentivized to invest ahead of demand, or if regulatory lag deters capital deployment, the pace and cost of decarbonization may be adversely affected. While yardstick regulation excels in steady-state environments with gradual change, it is less suited to contexts requiring anticipatory investment and network transformation. Indeed, the evidence of catch-up investment by DSOs post-2020 suggests that earlier regulatory assumptions underestimated the scale and speed of the transition. The analysis ultimately suggests that the regulatory model that served the sector and end-users well for much of the early 2000s must now evolve to meet new challenges. Yardstick regulation has delivered cost savings equal to 2.5 percent per year on a per-connection basis. However, tariffs have now started to rise to accommodate rising investments. The ACM has announced that it intends to move to a more forward-looking input-based model, capable of aligning revenue allowances with expected investments in grid flexibility, digitalization, and electrification. Stability and predictability in regulatory incentives are especially important in a sector that is simultaneously managing legacy infrastructure and preparing for a more dynamic, decentralized energy system. The question is whether with better anticipation the reversal in productivity could have been blunted and whether an earlier switch to a more forward-looking regulatory regime would have delivered higher outputs with less costs. Declarations Author Contribution P.N. wrote the main manuscript text and prepared the figures and tables. All authors reviewed the manuscript. References ACM factsheets, multiple years. ACM X-factorbesluiten RNB Elektriciteit, multiple regulatory periods. ACM (2012). De Toekomst van Tariefregulering , Report prepared for the ACM by PwC, 7 September 2012. ACM (2015). Rapportage winsten regionale netbeheerders elektriciteit en gas 2008–2013 , 31 maart 2015. ACM (2020). Frontier Shift for Dutch Gas and Electricity TSOs , Report prepared for the ACM by Economic Insights, 1 May 2020. ACM (2023). Toekomstige reguleringsmethode netbeheerders voor de reguleringsperiode vanaf 2027 , ACM/22/176591 / Documentnr. ACM/UIT/605442. AER (2024). Annual Benchmarking Report, Electricity Distribution Network Service Providers , November 2024. Ajayi, V., Dolphin, G., Anaya, K., & Pollitt, M. G. (2020). The Productivity Puzzle in Network Industries: Evidence from the Energy Sector, Working Papers EPRG 2021 , Energy Policy Research Group, Cambridge Judge Business School, University of Cambridge. Ajayi, V., Anaya, K., & Pollitt, M. G. (2022). Incentive regulation, productivity growth and environmental effects: the case of electricity networks in Great Britain, Energy Economics , Volume 115. https://doi.org/10.1016/j.eneco.2022.106354 Alliander (2009–2024), Annual Reports. Amundsveen, R., & Kvile, H. M. (2017). Productivity Development Among Norwegian Electricity Distribution System Operators - The Impact of Capital Assessment and Technological Progress. Data Envelopment Analysis Journal , 3 (1–2), 93–118. https://doi.org/10.1561/103.00000016 Baarsma, B., de Nooij, M., Koster, W., & van der Weijden, C. (2007). Divide and rule. The economic and legal implications of the proposed ownership unbundling of distribution and supply companies in the Dutch electricity sector. Energy Policy , 35 (3), 1785–1794. https://doi.org/10.1016/j.enpol.2006.05.002 Baumol, W. J. (1982). Productivity Incentive Clauses and Rate Adjustment for Inflation, Public Utilities Fortnightly , 110, July, pp. 11–18. Bruegel (2025). Upgrading Europe’s Electricity Grid is About More than Just Money , Bruegel Policy Brief 04/25, February 2025. CPB (2000). Yardstick competition: Theory, design, and practice , CPB Netherlands Bureau for Economic Policy Analysis, Working Paper, No. 133, December 2000, The Hague. CPB (2023). National Productivity Board 2022 Annual Report. DTe (1999). Price Cap Regulation in the Electricity Sector, Information and Consultation Document , July 1999, The Hague. Duma, D., Pollitt, M. G., & Covatariu, A. (2024). Regulatory Learning in the face of Net Zero Climate Policy: The Case of the UK. Review of Industrial Organization , 65 (2), 505–530. https://doi.org/10.1007/s11151-024-09972-y Eneco Holding (2013–2016), Annual Reports. Enexis (2015–2017), Reguleringsdata. Enexis Holding (2009–2024), Annual Reports. Enexis Netbeheer (2013–2024), Aanvulling segmentinformatie. Enexis Netbeheer (2009–2024), Annual Reports. European Heat Pump Association (2025). Heat pump sales 14 times greater in lead countries , https://www.ehpa.org/news-and-resources/news/heat-pump-sales-14-times-greater-in-lead-countries/ Eyre, S., & Pollitt, M. G. (2016). Introduction. In S. Eyre, & M. G. Pollitt (Eds.), Competition and Regulation in Electricity Markets (pp. xiii–xxxv). Edward Elgar. Fernandez, L. P., San Roman, T. G., Cossent, R., Domingo, C. M., & Frias, P. (2010). Assessment of the impact of plug-in electric vehicles on distribution networks. IEEE Transactions on Power Systems , 26 (1), 206–213. https://doi.org/10.1109/TPWRS.2010.2049133 Filippini, M., & Sánchez, L. O. (2014). Applications of the stochastic frontier approach in the analysis of energy issues. Economics and Business Letters , 3 (1), 35–42. https://doi.org/10.17811/ebl.3.1.2014.35-42 Financial Times (2025). Netherlands Rations Electricity to Ease Power Grid Stresses , 13 July 2025. Gupta, R., Pena-Bello, A., Streicher, K. N., Roduner, C., Farhat, Y., Thoeni, D., Patel, M. K., & Parra, D. (2021). Spatial analysis of distribution grid capacity and costs to enable massive deployment of pv, electric mobility and electric heating, Applied Energy , 287(C). https://doi.org/10.1016/j.apenergy.2021.116504 Gust, G., Schlüter, A., Feuerriegel, S., Úbeda, I., Lee, J. T., & Neumann, D. (2024). Designing Electricity Distribution Networks: The Impact of Demand Coincidence. European Journal of Operational Research , 315 (1), 271–288. https://doi.org/10.1016/j.ejor.2023.11.029 Haffner, R. C. G., & Meulmeester, P. (2005). Evaluatie van de Regulering van het Elektriciteitsnetwerk. Economische en Statistische Berichten , 4472 , 430–433. Hensgens, R., Nillesen, P. H. L., en, & Wessels, R. (2021). Energietransitie Vergt Aanpassing Regulering Netbeheerders. Economische en Statistische Berichten , 106 (4804), 578–581. Hinz, F., Schmidt, M., & Moest, D. (2018). Regional distribution effects of different electricity network tariff designs with a distributed generation structure: The case of Germany. Energy Policy , 113 , 97–111. https://doi.org/10.1016/j.enpol.2017.10.055 Hoefer, T., & Madlener, R. (2021). Locational (In) efficiency of renewable energy feed-in into the electricity grid: A spatial regression analysis. The Energy Journal , 42 (1). https://doi.org/10.5547/01956574.42.1.thof Jamasb, T., & Pollitt, M. G. (2001). Benchmarking and regulation: International electricity experience. Utilities Policy , 9 (3). https://doi.org/10.1016/S0957-1787(01)00010-8 . pp. 107 – 30. Joskow, P. L. (2014). Incentive Regulation in Theory and Practice: Electricity Distribution and Transmission Networks, in Rose, N.L. (Ed.), Economic Regulation and Its Reform: What Have We Learned? NBER. Joskow, P. L., & Schmalensee, R. (2025). Cost-of-service regulation of electricity distribution services in the US, in Glachant, J-M., Joskow, P.L., & Pollitt, M.G. (Eds.), Handbook on Electricity Regulation , Edward Elgar. Just, L., & Wetzel, H. (2020). Distributed Generation and Cost Efficiency of German Electricity Distribution Network Operators, EWI Working Paper , No. 20/09. Kendrick, J. W. (1975). Efficiency Incentives and Cost Factors in Public Utility Automatic Revenue Adjustment Clauses. Bell Journal of Economics , 6 (1), 299–313. Liander (2008). 2011–2024), Annual Reports. Littlechild, S. (1983). Regulation of British Telecommunications’ Profitability , Report to The Secretary of State, Department of Industry, February 1983. Lundgren, T., & Vesterberg, M. (2024). Efficiency in Electricity Distribution in Sweden and the Effects of Small-scale Generation, Electric vehicles and Dynamic tariffs. Journal of Productivity Analysis , 62 , 121–137. https://doi.org/10.1007/s11123-024-00724-4 McKinsey (2002). The Real Cost of Equity, Perspectives on Corporate Finance and Strategy , McKinsey on Finance, Number 5, Autumn 2002. Ministry of Economic Affairs (2003). Wet van 2 juli 1998, houdende regels met betrekking tot de productie, het transport en de levering van elektriciteit (Elektriciteitswet 1998) , (Stb. 2003, 235), The Hague. Montfoort van, F., Dijkstra, P. T., & Mulder, M. (2024). The impact of energy transition on distribution network costs and effectiveness of yardstick competition: an empirical analysis for the Netherlands. Journal of Regulatory Economics , 65 , 85–107. https://doi.org/10.1007/s11149-024-09471-8 Mulder, M., & Plug, P. (2009). Reguleren van energiemarkten: sturen in een dynamische omgeving in De blijvende uitdaging: Liber amicorum voor ir. Gert J.L. Zijl en drs. René Jansen , NMa Netherlands Competition Authority. Netbeheer Nederland (2024). Financiële Impact Energietransitie voor Netbeheerders (FIEN+) , PwC, 16 December 2024. Niesten, E. (2010). Network investments and the integration of distributed generation: Regulatory recommendations for the Dutch electricity industry. Energy Policy , 38 (8), 4355–4362. Nieuwenburg, M., Smit, E., & Mulder, M. (2020). Performance of Dutch energy distribution operators Performance of Dutch energy distribution operators an assessment of tariffs, quality and financial position since 2010 an assessment of tariffs, quality and financial position since 2010. Centre for Energy Economics Research (CEER) , Policy Papers 7. Nieuwsuur (2022). Reconstructie: Hoe Toezichthouder Tekorten op het Elektriciteitsnet Mede Veroorzaakte , maandag 3 oktober 2022 16.00. https://nos.nl/nieuwsuur/artikel/2446934 Nillesen, P. H. L., & Pollitt, M. G. (2007). The 2001-3 electricity distribution price control review in the Netherlands: regulatory process and consumer welfare. Journal of Regulatory Economics , 31 , 261–287. https://doi.org/10.1007/s11149-006-9014-8 Nillesen, P. H. L., & Pollitt, M. G. (2021). Ownership Unbundling of Electricity Distribution Networks. Economics of Energy and Environmental Policy , 10 (1), 147–158. https://doi.org/10.5547/2160-5890.10.1.pnil NMa. (2011). Onderzoek naar de methodologie voor de verdeling van de kosten van netverliezen (Vol. 29). KEMA and SEO. March 2011. PBL (2022). Maatschappelijke Kosten-Batenanalyse en Brede Welvaart; Een Aanvulling op de Algemene MKBA-Leidraad , June 2020. PBL (2024). Laadinfrastructuur mobiliteit, 2018–2023 . https:// monovi.pbl.nl/indicatoren/nl059902-laadinfrastructuur-mobiliteit-2018-2023#bronnen Pollitt, M. G., Duma, D., Covatariu, A., & Nillesen, P. H. L. (2025). A Global Map of Electricity and Gas Distribution Network Companies, Energy Policy Research Group, Working Papers No.2519, Cambridge Judge Business School, University of Cambridge. Ramos-Real, F. J., Tovar, B., Iootty, M., de Almeida, E. F., & PintoJr., H. Q. (2009). The Evolution and Main Determinants of Productivity in Brazilian Electricity Distribution 1998–2005: An Empirical Analysis, Energy Economics , vol. 31(2), pp. 298–305. https://doi.org/10.1016/j.eneco.2008.11.002 Regulatory Assistance Project (2019). Performance-based Regulation: Aligning Incentives with Clean Energy Outcomes , June 2019. RVO (2025). Charging and Tank Infrastructure in The Netherlands . https:// duurzamemobiliteit.databank.nl/mosaic/en-us/elektrisch-vervoer/laad--en-tankinfrastructuur-in-nederland Schleifer, A. (1985). A theory of yardstick competition. Rand Journal of Economics , 16(3). Solar Magazine (2025). De harde cijfers | SDE++: marktgroei 2024 bijgesteld tot bijna 2,2 gigawattpiek . https:// solarmagazine.nl/nieuws-zonne-energie/i40177/de-harde-cijfers-sde-marktgroei-2024-bijgesteld-tot-bijna-2-2-gigawattpiek Stedin (2016–2024), Annual Reports. Stedin Groep (2016–2024), Annual Reports. Tertre (2023). Structural Changes in Energy Markets and Price Implications: Effects of the Recent Energy Crisis and Perspectives of the Green Transition , European Central Bank. Tweede Kamer der Staten-Generaal. (2005). Wijziging van de Elektriciteitswet 1998 en van de Gaswet in verband met nadere regels omtrent een onafhankelijk netbeheer (Vol. 30212). nr. 3, Memorie van Toelichting. Vlijm, W. (2002). De Interactie tussen de Overheid en de Elektriciteitssector in Nederland: De Ontwikkeling van het Nutsbedrijf PGEM naar de Energieonderneming Nuon (1916–2001) , PhD Dissertation, Radboud Universiteit. Vogelsang, I. (2002). Incentive Regulation and Competition in Public Utility Markets: A 20-Year Perspective. Journal of Regulatory Economics , 22 , 5–27. https://doi.org/10.1023/A:1019992018453 Wangsness, P. B., & Halse, A. H. (2021). The impact of electric vehicle density on local grid costs: Empirical evidence from Norway. The Energy Journal , 42 (5). https://doi.org/10.5547/01956574.42.5.pwan Footnotes Although it was actually first suggested for US telecoms by Baumol ( 1982 ) drawing on ideas in Kendrick ( 1975 ), as discussed in Eyre and Pollitt ( 2016 ). For a recent discussion see Joskow and Schmalensee ( 2025 ). Ministry of Economic Affairs ( 2003 a), Wet van 2 juli 1998, houdende regels met betrekking tot de productie, het transport en de levering van elektriciteit (Elektriciteitswet 1998) (Stb. 2003, 235), The Hague. In Dutch: Samengestelde Output (SO). Following the first Oil Crisis in 1973, Dutch government policy was focused on getting more grip on the energy sector and energy provision. It was active government policy to consolidate the energy sector in the Netherlands, most notably with the report from the Commissie Concentratie Nutsbedrijven (CoCoNut) in 1980 and the report from the Commissie Brandsma in 1985. In 1985 there were 158 energy companies. The recommendations were to consolidate this to 56 companies in 1987 and 48 in 1991 (see Vlijm, 2002 ). The three smaller DSOs are Coteq (0.6 percent market share), Rendo (0.4 percent market share), and Westland Infra (0.7 percent market share). The three DSOs that were consolidated into three large DSOs were Enduris, NRE, and ONS. There have also been several service area swaps between the three large DSOs. IFRS = International Financial Reporting Standards. In the first regulatory period 2001–2003, the ACM determined the starting value for the RAB (start-GAW), which the ACM treats as an investment done at the end of 2000, with one uniform depreciation period. We discuss this restructuring in Nillesen and Pollitt ( 2021 ). We base the one-off unbundling costs on the submission to Parliament by the Ministry of Economic Affairs in 2005, which provides the financial impact assessment of Deloitte (Tweede Kamer der Staten-Generaal, 2005 ). The costs associated with unbundling (one-off unbundling costs estimated between €80-130mn) were not allowed to be passed through to end-users by law. We assume that all three DSOs incurred the costs in 2009, even though Stedin and Eneco unbundled later following court appeals. NMa ( 2011 ) estimates that losses cost approximately EUR 30 per customer and is equivalent to ~ 5 percent of total distributed volume (70 percent technical and 30 percent administrative) of which 82 percent occurs at MS/LS level. The study projects the total investment in infrastructure for electricity (transmission and distribution), gas (transport and distribution), heat networks, and hydrogen infrastructure to be approximately €200bn up to 2040 (Netbeheer Nederland 2024 ). On a per capita basis, the Netherlands has the highest solar PV penetration in the world. There are 614k full electric vehicles and 458k plug-in hybrid vehicles as per July 2025, according to CBS. See European Heat Pump Association, 3 July 2025. See for example, Hinz et al. ( 2018 ), Hoefer & Madlener ( 2021 ), Halse (2021), and Filipini & Sanchez (2014). This could have implications for the effectiveness of yardstick composition, as distribution costs increase but the output (volume per connection) remains constant. If different DSOs are exposed to different degrees of energy transition-related investments in the network, the outcome from yardstick competition could lead to over or under compensation for individual DSOs. They observe 0–10 percent changes in individual price caps for most DSOs, and even an increase of 20 percent for one of the DSOs. SAIDI = System Average Interruption Duration Index. See Pollitt et al. ( 2025 ) where all the DSOs from capitals around the world are compared. The effective X-factor for the first regulatory period was 2.26 percent, although the reported X-factor that was negotiated was 3.2 percent. This is the result of the no “reformatio in peius” principle, which meant that decisions by the regulator, if more favourable, could not be changed. The third regulatory period was shortened to one year (instead of three years as initially planned), given the transfer of the 110/150kV lines from the DSOs to TenneT and the change to yardstick regulation (as opposed to the individual benchmarking in the first and second regulatory period). The negotiated outcome for the third period was an overall tariff freeze in reals terms for electricity and gas combined. If we assume that unit operational costs develop in line with TFP for the Netherlands, then the benefits of yardstick regulation decline by approximately €500mln to €2.1bn. McKinsey (2002) analysis suggests that investors seek a stable total return and that during periods of low interest rates the market risk premium rises and vice versa . This suggests that the equity component of the WACC remains stable over time, rather than vary with the risk-free rate of interest. They estimate the average inflation-adjusted cost of equity implied by stock market valuations each year from 1963 to 2001 in the US and from 1965 to 2001 for the UK to be 7 percent. Our estimate using the average from the ACM is 7.6 percent and in line with the McKinsey estimates. Tables Tables 1 to 5 are available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files Table1.Descriptionofvariablesandsources.jpg Table2.Collecteddata20002024.jpg Table3.Summaryofmainresults.jpg Table4.ChangeinunderlyingcomponentsoftheConstructedRegulatedRevenueperRegulatoryPeriod.jpg Table5.ChangeinConstructedRegulatedRevenueperConnectionperRegulatoryPeriod.jpg Cite Share Download PDF Status: Published Journal Publication published 26 Apr, 2026 Read the published version in Journal of Regulatory Economics → Version 1 posted Editorial decision: Revision requested 25 Sep, 2025 Editor assigned by journal 17 Sep, 2025 Submission checks completed at journal 17 Sep, 2025 First submitted to journal 12 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7601011","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":520748193,"identity":"ff51ab28-7398-424a-b9cf-1077913bc529","order_by":0,"name":"Paul H L Nillesen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2UlEQVRIiWNgGAWjYHACxgMMDBIyfGB2BZF6QFp42MDMM8RrYYBoYWwjQrk5+xmDA4w7LHjY2NsfPi6cdyexv4H54aMbeLRY9uQAtZwBOoznjLHxzG3PEmccYDM2zsGjxeAASEsbUItEDps077bDuRuAjpTGq+X8G6gW+efPf/POIUbLDbgtDGbMvA1EaLGc8azgQCJIC0+OsTTPsWf1Mw4T8Is5f/LGBx/b6uT42Y8//MxTc8eYv7354WO8DgMRCQj+AQYGZjzK4VqQwAEC6kfBKBgFo2AkAgC4W0cYXYPZDwAAAABJRU5ErkJggg==","orcid":"","institution":"Partner PricewaterhouseCoopers Advisory N.V. Thomas Malthusstraat Amsterdam","correspondingAuthor":true,"prefix":"","firstName":"Paul","middleName":"H L","lastName":"Nillesen","suffix":""},{"id":520748194,"identity":"b5935e25-8df6-44ca-b17a-321d0895d312","order_by":1,"name":"Michael G Pollitt","email":"","orcid":"","institution":"Cambridge University","correspondingAuthor":false,"prefix":"","firstName":"Michael","middleName":"G","lastName":"Pollitt","suffix":""}],"badges":[],"createdAt":"2025-09-12 13:23:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7601011/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7601011/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s11149-026-09511-5","type":"published","date":"2026-04-26T15:58:53+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":93731313,"identity":"2cea561a-2d66-40f1-bb18-9843c7acdd05","added_by":"auto","created_at":"2025-10-17 02:23:30","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":284990,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eConstructed Regulated Revenue per Connection, 2000-2024 (real 2000)\u003c/em\u003e\u003c/p\u003e","description":"","filename":"Figure1ConstructedRegulatedRevenueperConnection20002024real2000.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7601011/v1/a1e8f79633e1d9a9eadddcdb.jpg"},{"id":93731358,"identity":"15baf6ac-9ee6-41a7-aa76-a97796824236","added_by":"auto","created_at":"2025-10-17 02:23:32","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":325547,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eConstructed Regulated Revenue components, indexed (2000=100)\u003c/em\u003e\u003c/p\u003e","description":"","filename":"Figure2ConstructedRegulatedRevenuecomponentsindexed2000100.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7601011/v1/be256d9eea318a6b0c0c3155.jpg"},{"id":93731319,"identity":"4612230b-46b9-4131-9d63-4ed8397daecd","added_by":"auto","created_at":"2025-10-17 02:23:30","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":330040,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eAverage return per Connection and the Real pre-tax WACC\u003c/em\u003e\u003c/p\u003e","description":"","filename":"Figure3AveragereturnperConnectionandtheRealpretaxWACC.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7601011/v1/523043e2ae81cb6a35eb0dac.jpg"},{"id":93731287,"identity":"037965e0-fe1f-4902-b511-d4dcb0153586","added_by":"auto","created_at":"2025-10-17 02:23:29","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":253969,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eChange in fixed assets (real)\u003c/em\u003e]\u003c/p\u003e","description":"","filename":"Figure4ChangeinFixedAssetsreal.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7601011/v1/6630c8308bd1a7492da9dc21.jpg"},{"id":93731288,"identity":"fb4ba2a3-3090-4646-a9fd-47dabd38bccc","added_by":"auto","created_at":"2025-10-17 02:23:29","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":472053,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eFixed assets, feedin capacity (solar and wind onshore) and off-take capacity (charge point and heat pump) over time\u003c/em\u003e\u003c/p\u003e","description":"","filename":"Figure5fixedassetsRHSfeedincapacitysolarandwindonshoreLHSandofftakecapacityEVchargeinfraandheatpumpLHSovertime.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7601011/v1/d0012fd11324168eb12fafd2.jpg"},{"id":93731276,"identity":"f1651b1a-3c94-4c12-90bf-64cca97ffe0f","added_by":"auto","created_at":"2025-10-17 02:23:27","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":483015,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eProductivity growth indexed (2000 = 100)\u003c/em\u003e\u003c/p\u003e","description":"","filename":"Figure6Productivitygrowthindexed2000100.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7601011/v1/5b32ceb201ebd268fc129e27.jpg"},{"id":93731312,"identity":"146a2554-6d16-4b3f-8e78-2aad7187a61a","added_by":"auto","created_at":"2025-10-17 02:23:30","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":435014,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eConstructed Regulated Revenue vs. ACM Projected Revenue (index 2000=100)\u003c/em\u003e\u003c/p\u003e","description":"","filename":"Figure7ConstructedRegulatedRevenuevs.ACMProjectedRevenueindex2000100.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7601011/v1/c24085b0c2df0efcffa9fb90.jpg"},{"id":107928103,"identity":"20be3dba-05ee-45a6-aa60-3082d2f8dd1a","added_by":"auto","created_at":"2026-04-27 16:07:55","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2942146,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7601011/v1/4022a5e0-b651-4ba2-b816-66b7acf6e6c1.pdf"},{"id":93731330,"identity":"76ae04c0-00a3-4b93-b2e4-739b5e07fedc","added_by":"auto","created_at":"2025-10-17 02:23:31","extension":"jpg","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":679961,"visible":true,"origin":"","legend":"","description":"","filename":"Table1.Descriptionofvariablesandsources.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7601011/v1/bc585bb7bf9e40d68a0b37e9.jpg"},{"id":93731277,"identity":"e7618ca8-075d-4539-95ac-7cfa3ab46991","added_by":"auto","created_at":"2025-10-17 02:23:27","extension":"jpg","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":1162937,"visible":true,"origin":"","legend":"","description":"","filename":"Table2.Collecteddata20002024.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7601011/v1/bbd2cab51fc34671b17220be.jpg"},{"id":93731323,"identity":"8fbf147d-765e-4d1e-a274-1e2f8af13b1d","added_by":"auto","created_at":"2025-10-17 02:23:31","extension":"jpg","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":821156,"visible":true,"origin":"","legend":"","description":"","filename":"Table3.Summaryofmainresults.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7601011/v1/ed37054753917c492fc9ba75.jpg"},{"id":93731284,"identity":"732fee17-0340-45c8-9598-4849cb324210","added_by":"auto","created_at":"2025-10-17 02:23:29","extension":"jpg","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":590179,"visible":true,"origin":"","legend":"","description":"","filename":"Table4.ChangeinunderlyingcomponentsoftheConstructedRegulatedRevenueperRegulatoryPeriod.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7601011/v1/aacf29eb3783193ab59b5a25.jpg"},{"id":93731279,"identity":"7d76d854-d9ae-4fe1-9731-3131ffb50ac3","added_by":"auto","created_at":"2025-10-17 02:23:28","extension":"jpg","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":631698,"visible":true,"origin":"","legend":"","description":"","filename":"Table5.ChangeinConstructedRegulatedRevenueperConnectionperRegulatoryPeriod.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7601011/v1/79e56cee0b60ad9ea4c7f160.jpg"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eIncentive Regulation and Distribution Network Performance: A Dutch Case Study (2000–2024)\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eThe Netherlands adopted a price control review approach to the economic regulation of its electricity Distribution System Operators (DSOs) in 2000. The Dutch DSOs have been subject to regulatory overview and yardstick competition for 25 years, based on an \u003cem\u003eex-ante\u003c/em\u003e price cap with an exogenous efficient cost level. Yardstick competition ensures that during the regulatory period all efficient costs at sector-level are recovered through tariffs. The regulatory objectives of the national energy regulator (ACM) over the full period have remained in place: (i) provide network operators an incentive to operate in an efficient manner, (ii) prevent network operators from charging tariffs above the (efficient) cost level, (iii) allow network operators to earn an appropriate return on invested capital, and (iv) encourage optimal quality of transport.\u003c/p\u003e\u003cp\u003eThere have been a number of studies examining the impact of regulation on costs, quality and investments in The Netherlands. For example, Nillesen \u0026amp; Pollitt (\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2007\u003c/span\u003e) analyse the impact on consumer welfare from the failed first distribution price control review. Baarsma et. al. (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2007\u003c/span\u003e) discuss the economic and legal implications of ownership unbundling of distribution networks from supply companies. Niesten (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) makes regulatory recommendations to facilitate the integration of distributed generation. Nieuwenburg et. al. (\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) assess the financial performance of network operators between 2012 and 2019. Montfort \u003cem\u003eet. al.\u003c/em\u003e (2024) analyse the impact of regional differences related to the energy transition on yardstick regulation outcomes between 2012 and 2020.\u003c/p\u003e\u003cp\u003eThis paper takes a longer-term perspective by presenting a detailed empirical assessment over a 25-year period. It examines the question how incentive regulation \u0026ndash; particularly yardstick benchmarking \u0026ndash; has affected the financial, operational, and investment performance of electricity DSOs in the Netherlands from 2000 to 2024. The analysis is timely from both a policy and regulatory perspective given the focus on electrification and the concurrent investments that are required to meet increasing demand because of the energy transition. The focus of the past 25 years has been on lowering end-user tariffs through efficiency incentives. As more distributed renewable energy has been connected to the network and more demand is being electrified (such as the electrification of transport), network investments have been increasing substantially in the past few years and are projected to rise significantly in the next decades (see Netbeheer Nederland, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). The rapid electrification in the Netherlands has also meant that investments have lagged demand, which has resulted in congestion and significant grid constraints (see for example, Financial Times, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Further, the ACM has announced that it is considering changing the regulatory framework from 2027 onwards, from the current output-based yardstick competition system to a more forward-looking input-based system, to account for increasing investments and the need for more forward-looking guidance (ACM, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). This would bring Dutch regulation more in line with the forward-looking practices implemented for energy network price controls in Great Britain by Ofgem (see Duma et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe structure of the paper is as follows. Section 2 outlines the regulatory context. Section 3 gives an overview of the data and the revenue proxy we construct. Section 4 provides the results from the analysis. Section 5 explores the limitations of backward-looking regulation and draws policy lessons.\u003c/p\u003e"},{"header":"2. Incentive Regulation in The Netherlands","content":"\u003cp\u003eIncentive-based regulation is commonly referred to as RPI-X (or CPI-X) regulation outside the US and was first suggested by Littlechild (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e1983\u003c/span\u003e).\u003csup\u003e1\u003c/sup\u003e This form of regulation allows average tariffs (0r revenues) to increase inline with inflation less an efficiency factor (the X-Factor), which reflects the potential for relative productivity improvements. The linking of regulated prices to performance is known as Performance-based Ratemaking (PBR) in the US.\u003csup\u003e2\u003c/sup\u003e Under both PBR and RPI-X regulated firm retain extra profits for the duration of the regulatory period. This form of price-cap regulation thus decouples profits from costs by setting maximum average tariffs (or revenues), by creating a regulatory lag (see Vogelsang, \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2002\u003c/span\u003e and Joskow, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eTo set the appropriate X-Factor or performance metric, many regulators have applied benchmarking techniques to compare the regulated firm\u0026rsquo;s actual performance against some pre-defined reference or benchmark performance. As highlighted by Schleifer (\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e1985\u003c/span\u003e) this approach has the theoretical advantage that it reduces the informational asymmetry that occurs in rate of return (ROR) regulation by reducing the regulator\u0026rsquo;s reliance on the firm\u0026rsquo;s own costs. The most disputatious part of price-cap regulation is the benchmarking technique used to determine the productivity gap and how outcomes are translated into X-Factors. Regulators have adopted a variety of benchmarking methods to arrive at X-Factors. There are broadly three different approaches: (i) historically observed costs (US ROR regulation), (ii) projections about firm-specific efficiency (European and Australian regulators), and (iii) industry-wide (or wider economy) productivity indices.\u003c/p\u003e\u003cp\u003eEuropean regulators have generally adopted firm-specific projection-based benchmarking methods to calculate X-factors, such as Data Envelopment Analysis using international data from other regulated entities. PUCs that adopted PBR have tended to use either historically observed costs (ROR) or industry wide productivity measures such as Total Factor Productivity (TFP) to calculate the efficiency requirements (see Jamasb and Pollitt, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2001\u003c/span\u003e). One of the main advantages of this approach is that it does not involve a direct comparison of the regulated firm with another firm\u0026rsquo;s costs. This can expose the regulator to legal challenges based on unfairness given incompatibility (as documented in Nillesen and Pollitt, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2007\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe Netherlands adopted a price control review approach to economic regulation in 1998 with the Electricity Act.\u003csup\u003e3\u003c/sup\u003e Originally the first price control period was to be from 1 January 2001 to 31 December 2003. The first consultation document for this price control review period was published in July 1999 (DTe, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e1999\u003c/span\u003e). The outcome of the first regulatory period was disappointing. In terms of the length of overrun in the process (the first X-Factor was set in summer of 2000, but final X-Factors were only agreed in May 2003) and in the unmet initial expectations, it compares unfavourably to other price reviews that regularly take place in for example the UK, Norway, Australia, and Chile. Of the initially promised savings of \u0026euro;511mln only \u0026euro;384mln turned out to be accurate. The subsequent court case defeats (for the regulator) and previous allowance of the \u0026ldquo;\u003cem\u003eno reformatio in peius\u0026rdquo;\u003c/em\u003e meant that only \u0026euro;209mln of the \u0026euro;384mln was achieved.\u003c/p\u003e\u003cp\u003eThe ACM subsequently abandoned individual benchmarking and introduced weighted-average yardstick competition based on Schleifer (\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e1985\u003c/span\u003e) in the second regulatory period (from 2004), where network operators are compared to mimic competition on cost levels (CPB, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). This approach is therefore less about finding theoretical \u0026ldquo;efficient costs\u0026rdquo;, but more about creating incentives to lower costs in a repeated game.\u003c/p\u003e\u003cp\u003eThe approach chosen by ACM attempts to capture the output generated by the DSO and when combined, the sector. The DSO supplies a portfolio of products and services, such as connection services, transportation services, and feed-in services, each with an associated tariff. The multiplication of the quantities of these products and services with the respective tariffs gives a \u0026ldquo;composite output\u0026rdquo; for each DSO.\u003csup\u003e4\u003c/sup\u003e The aggregate output, for the whole sector, is then used to calculate average sector costs. This is simply defined as the total costs of all DSOs (based on a TOTEX approach) divided by the total aggregate output of all DSOs. In principle all operational costs are included in the yardstick including network losses (technical and administrative), except for transmission costs (TenneT) and transmission costs from other DSOs. Depreciation and the allowed return (WACC) are based on the Regulated Asset Base (RAB). The sum of operational costs, depreciation (based on the RAB), and the allowed return (WACC * RAB), is the TOTEX for the sector. At the start of each reporting period, the RAB is updated to account for depreciation and investments.\u003c/p\u003e\u003cp\u003eThe sector-weighted average cost per unit of output forms the basis for the yardstick against which the performance of individual DSOs is assessed and used to determine the regulated revenue and tariffs. Dynamic efficiency is introduced by extrapolating historic productivity growth. As such, the combined costs of the sector are recouped through tariffs within a regulatory period, although an individual DSO can earn a surplus or have a deficit if their individual unit costs are lower or higher respectively. For certain cost components (e.g. transmission costs or network losses) the ACM has announced that it will allow recovery earlier in the regulatory period. For example, transmission costs from TenneT are difficult to forecast. In the past these costs were determined ex-ante with a two-year delay for determining actual costs, leading to significant pre-financing by the DSOs (without any compensation). Other costs, such as investments related to the energy transition and the feed-in of decentralized generation will be determined ex-post.\u003c/p\u003e\u003cp\u003eIn the past there have been numerous legal challenges to price control reviews. Some of these challenges have resulted in adjustments to the methodology or the interpretation of the exact application of the methodology. This means that in certain years or regulatory periods there is the possibility of under or over recovery, and subsequent adjustments in the next price control review.\u003c/p\u003e"},{"header":"3. Methodology and Data","content":"\u003cp\u003eThe Netherlands currently has six electricity DSOs. Three of which (Liander, Enexis, and Stedin) account for more than 98 percent of all connections. At the start of regulation in 2000 there were nine electricity DSOs. Three of the smaller DSOs were acquired by the big three players during the period we analyse. At the start of regulation, in 2000, the big three DSOs had a combined market share of just under 94 percent. The big three DSOs themselves are the result of multiple mergers between smaller utility companies.\u003csup\u003e5\u003c/sup\u003e Although separately benchmarked initially, they were subsequently viewed as single players by the regulator as operations, investment, and management were consolidated.\u003csup\u003e6\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eTo assess the financial and operational performance of the sector over the last 25 years we have collected data from various sources. Given the timespan under consideration there is no single data source. We have therefore created a dataset using multiple sources, such as company annual reports and company publications, regulatory filings and regulatory publications, and publicly available information. The data is based on the three largest DSOs \u0026ndash; Liander, Enexis, and Stedin. Table\u0026nbsp;1 provides a description of the variables.\u003c/p\u003e\u003cp\u003e[\u003cem\u003eTable\u0026nbsp;1 \u0026ndash; description of variables sources\u003c/em\u003e]\u003c/p\u003e\u003cp\u003eWe construct a proxy for the sector-level regulated revenue \u0026ndash; the Constructed Regulated Revenue \u0026ndash; by applying the building block approach. This includes operational costs, transmission costs (TSO cost pass-through), network losses, depreciation, and the regulated return on invested capital. See formula 1.\u003c/p\u003e\u003cp\u003e[\u003cem\u003eFormula 1 \u0026ndash; Constructed Regulated Revenue\u003c/em\u003e]\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:{Constructed\\:Regulated\\:Revenue}_{i,t}={Operating\\:Costs}_{i,t}+\\:{Transmission\\:Costs}_{i,t}+\\:{Network\\:Losses}_{i,t}+\\:{Depreciation}_{i,t}+\\:{Return}_{i,t}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cem\u003eWhere i is the DSO and t the year. The Return is calculated using formula 2.\u003c/em\u003e\u003c/p\u003e\u003cp\u003e[\u003cem\u003eFormula 2 \u0026ndash; Return\u003c/em\u003e]\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\:{Return}_{i,t}={Real\\:pre\\text{-}tax\\:WACC}_{t}*{Fixed\\:Assets}_{i,t}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cem\u003eWhere i is the DSO and t the year.\u003c/em\u003e\u003c/p\u003e\u003cp\u003eWhere possible we use data directly from annual reports or regulatory filings. Where data is not available or missing, we interpolate or make estimates based on reference figures. For example, in some years transmission costs and network losses are not reported separately, but as one category. Another challenge is the fact that there is limited data available for the DSOs prior to the compulsory ownership unbundling, when data was aggregated at group levels.\u003c/p\u003e\u003cp\u003e[\u003cem\u003eTable\u0026nbsp;2 \u0026ndash; Collected data 2000\u0026ndash;2024\u003c/em\u003e]\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;2 provides an overview of the collected data. This includes the financial variables related to the Constructed Regulated Revenue, such as operational costs, fixed assets, and network losses. We also present operational statistics, such as number of connections, network length, volumes transported, and SAIDI. Table\u0026nbsp;2 also reports data related to the regulatory framework, such as the pre-tax real WACC, CPI, the regulatory period, and the connection-weighted average X-factor per regulatory period. Finally, we present related data for the analysis, such as wholesale electricity prices, Total Factor Productivity for the Netherlands, electricity demand, installed capacity of solar PV and onshore wind, and the number and estimated capacity of EV charging stations.\u003c/p\u003e\u003cp\u003eThe Constructed Regulated Revenue is compared with the actual or estimated reported revenue. Over the full 25 years the total difference between our estimate of regulated revenue and the reported revenue is negligible (less than 0.5 percent). However, in certain years there are differences of more than 5 percent. Thus, in some years we are over- or underestimating the actual regulated revenue. This is due to several factors. In some years the data is incomplete or missing (leading to assumed values) or the level of reporting is different (e.g. at group level versus at DSO level). There are timing effects we do not consider. For example, certain costs, such as transmission charges are based on regulatory estimates with actual costs recouped in later years. In addition, there have been numerous legal proceedings that have resulted in adjustments and ex-post corrections to correct for over or under recovery across regulatory periods. Our estimate for the allowed return uses the reported asset value in the annual report (based on IFRS\u003csup\u003e7\u003c/sup\u003e) as a proxy for the RAB. The regulatory accounting rules deviate in certain aspects from IFRS. In addition, at the start of the first regulatory period the RAB was treated as one asset class with one remaining depreciation period.\u003csup\u003e8\u003c/sup\u003e Over time the RAB value therefore deviates from the underlying IFRS-based valuation of the invested capital.\u003c/p\u003e\u003cp\u003eThe main deviations seem to occur during periods where the projected development of costs by the regulator deviate from the actual development, or when structural changes occur. An example of this is in 2020/21 at the end of the seventh regulatory period, when the projections for 2022\u0026ndash;2026 were made. This did not include the severe impact on wholesale energy prices from the Russian invasion of Ukraine. Wholesale electricity prices were more than three times higher in 2021 than in 2020, and more than doubled again in 2022. This had a notable effect on the value of network losses, which were significantly higher than estimated in 2020/21. This leads to a significant difference between our Constructed Regulated Revenue and the reported revenue based on tariffs set by the regulator. For example, our Constructed Regulated Revenue for 2022-24 is approximately 6 percent higher.\u003c/p\u003e\u003cp\u003eIn general, these deltas between actual and projected revenue are corrected and recouped in subsequent regulatory periods. The other reason for significant deviations is structural in nature.\u003csup\u003e9\u003c/sup\u003e Liander and Enexis were fully unbundled at ownership level from NUON and Essent respectively in 2009. Stedin was only fully unbundled from Eneco in 2017. Although we have attempted to correct for this based on publicly-available information, it is likely that the actual costs deviate in magnitude and timing.\u003csup\u003e10\u003c/sup\u003e In addition, in 2008 the 110/150kV networks owned by the DSOs were transferred to TenneT, the TSO. This is also one of the reasons why the third regulatory period was shortened to one year instead of three years.\u003c/p\u003e\u003cp\u003eThe financial data we present and use for our analysis therefore provides a directional view of the impact of incentive-based yardstick regulation on the financial performance of the sector as a whole and smoothes out the ex-post calculation effects. The Constructed Regulated Revenue is therefore a proxy of the average connection costs borne by customers connected to the distribution grid. It represents the \u0026ldquo;average network bill\u0026rdquo; for end-users (as represented by connections). In this case the end-user is not just a typical household but includes connections to small and medium enterprises and separately-metered EV charge points for example.\u003c/p\u003e\u003cp\u003eGiven the fact that there is no single data source and our dataset comprises of multiple sources, with some estimations and interpolations when data is missing, we only examine the development of the financial and operational performance at sector-level and do not compare the performance of Liander, Enexis, and Stedin with each other.\u003c/p\u003e"},{"header":"4. Results","content":"\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cem\u003eAverage network costs over time\u003c/em\u003e\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cp\u003eTo compare the results over time, we deflate the financial data to 2000 prices using the Consumer Price Index (CPI). Table\u0026nbsp;3 summarises the main results from our analysis and Fig.\u0026nbsp;1 shows the development of the Constructed Regulated Revenue per connection from 2000 to 2024.\u003c/p\u003e\u003cp\u003e[\u003cem\u003eTable\u0026nbsp;3 \u0026ndash; Summary of main results\u003c/em\u003e]\u003c/p\u003e\u003cp\u003e[\u003cem\u003eFigure 1 - Constructed Regulated Revenue per Connection, 2000\u0026ndash;2024 (real 2000)\u003c/em\u003e]\u003c/p\u003e\u003cp\u003eThe Constructed Regulated Revenue per connection rose from \u0026euro;284 in 2000 to \u0026euro;381 in 2024, a real-term increase of 1.2 percent per year. This growth, however, masks substantial heterogeneity: from 2000 to 2022, average costs declined by 0.8 percent annually, but surged post-2021 due to two primary drivers. First, the energy crisis following Russia\u0026rsquo;s invasion of Ukraine drastically increased wholesale prices, affecting network losses (see for example Tertre, 2023). Second, TSO transmission costs tripled between 2021 and 2024 due to elevated investment levels. In 2022 the average wholesale electricity price in the Netherlands was eight times higher than the price in 2020. This commodity cost flows into the costs for end-users through the cost for compensating network losses (which are approximately 4.5 percent of transported volumes at the distribution level\u003csup\u003e11\u003c/sup\u003e) at both the DSO level and via transmission charges from the TSO. Losses per connection were twice as high as average in 2023 for example. Second, transmission costs \u0026ndash; which are passed-through \u0026ndash; have increased due to increasing investment levels by the TSO. Transmission costs per connection in 2024 are almost three times higher than in 2021. In the period up to 2022 average connection costs decreased by 0.8 percent per year or 14.9 percent. Excluding the pass-through costs from the TSO gives a view on the core DSO-related costs. Over the full period the Constructed Regulated Revenue per connection decreased by 0.4 percent per year. The average distribution network connection costs were approximately 10 percent less in 2024 than in 2020 (in real terms).\u003c/p\u003e\u003cp\u003e[\u003cem\u003eFigure 2 - Constructed Regulated Revenue components, indexed (2000\u0026thinsp;=\u0026thinsp;100)\u003c/em\u003e]\u003c/p\u003e\u003cp\u003eOperational costs per connection declined by 21.6 percent between 2000 and 2018. Since 2018 operational costs have however been increasing. In 2023 operational costs per connection were the same as at the start of regulation, and in 2024 14.3 percent higher than in 2000. Operational costs are strongly related to the size of the asset base, the level of investment, and the share of distributed generation in the network. We discuss the impact of increasing investment levels later.\u003c/p\u003e\u003cp\u003eThe most striking change over the full period is the drop in the allowed return per connection \u0026ndash; almost halving from \u0026euro;83.4 per connection in 2000 to \u0026euro;42.8 in 2024. This significant decrease is the result of a combined effect of a lower allowed rate of return by the regulator and higher capital utilization. The regulatory real pre-tax WACC dropped from 6.7 percent in 2000 to 3.5 percent in 2024, as interest rates declined over this period (see Fig.\u0026nbsp;3).\u003c/p\u003e\u003cp\u003e[\u003cem\u003eFigure 3 - Average return per Connection and the Real pre-tax WACC\u003c/em\u003e]\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cem\u003eInvestments and the energy transition\u003c/em\u003e\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cp\u003eAt the same time capital intensity increased between 2000 and 2017, with fixed assets per connection dropping by 25 percent. This improvement in capital efficiency \u0026ndash; adding connections at a higher rate than expanding the capital base \u0026ndash; further put downward pressure on the allowed return per connection. From 2018 onwards average fixed assets per connection have increased by more than 31 percent, with the average fixed assets per connection in 2024 equal to the level at the start of regulation in 2000. Overall, the impact on the total allowed return of the increase in the asset base is outweighed by the significant drop in the allowed return (+\u0026euro;155mln versus -/-\u0026euro;349mln in allowed return, respectively).\u003c/p\u003e\u003cp\u003e[\u003cem\u003eFigure 4 \u0026ndash; Change in fixed assets (real)\u003c/em\u003e]\u003c/p\u003e\u003cp\u003eFigure 4 shows the change in fixed assets over time. Following a period up to 2008 where depreciation exceeded investment levels and the asset base declined, there was a period between 2009 and 2016 when the asset base expanded by circa \u0026euro;190mn per year. From 2017 investment levels have been increasing year-on-year. Between 2023 and 2024 alone, the three DSOs invested \u0026euro;2.3bn. The Capex-to-Opex ratio also shows how investment levels are increasing. At the start of regulation, the ratio was less than 0.5 but has been close to or greater than 1 since 2019. Projections by the association of network operators (Netbeheer Nederland) show that investment levels will increase further in the next decades because of increased electrification and higher renewable penetration.\u003csup\u003e12\u003c/sup\u003e\u003c/p\u003e\u003cp\u003e[\u003cem\u003eFigure 5 - Fixed assets, feedin capacity (solar and wind onshore) and off-take capacity (charge point and heat pump) over time\u003c/em\u003e]\u003c/p\u003e\u003cp\u003eFigure 5 shows the growth in renewable feed-in capacity (solar PV and onshore wind) and the growth in energy transition off-take capacity (charge point infrastructure and heat pumps) compared with the asset base of the three large DSOs in the Netherlands. Wind onshore installed capacity is currently 7GW (12 percent CAGR 2000-24). Solar PV has been even more successful, with an installed capacity of 30GW.\u003csup\u003e13\u003c/sup\u003e The installed capacity of solar PV in 2024 (23.4GW) was almost twenty times higher than a decade earlier in 2014 (1GW). The growth in distributed generation accelerated in 2017. The annual growth rate between 2000 and 2016 was 13.6 percent, whereas between 2017 and 2024 this almost doubled to 25.6 percent per year.\u003c/p\u003e\u003cp\u003eThe growth in EV charge points shows a similar pattern, following the sharp growth in the number of EVs.\u003csup\u003e14\u003c/sup\u003e The annual growth rate from 2015 \u0026ndash; when the first reliable data is available \u0026ndash; is more than 33 percent. There are currently almost 1mn charge points in the Netherlands. This includes home charging, public charging, semi-public charging, office charging, and high-powered charging. We present data that excludes home and office charge points, given that these connections do not require grid reinforcements and are \u0026ldquo;behind-the-meter\u0026rdquo;. Using average capacities for each category, we estimate the total installed capacity of these charge points. This has grown from 337MW in 2015 to 3.5GW in 2024.\u003c/p\u003e\u003cp\u003eA further example of the electrification of energy demand that impacts electricity network operators is the switch from natural gas to heat pumps for heating. There has been significant growth in the number of installed heat pumps (air and water-based). From 49 thousand units in 2015, there are now 2.3mn units installed \u0026ndash; an annual growth rate of more than 50 percent. This is still less than half compared to Europe\u0026rsquo;s leading country Norway with 632 per 1000 households.\u003csup\u003e15\u003c/sup\u003e The installed capacity of these heat pumps has grown from 354MW in 2015 to more than 13GW in 2024.\u003c/p\u003e\u003cp\u003eThe energy transition is responsible for the sharp increase in investment levels by the DSOs. It is not just direct and indirect investments in the network infrastructure, such as new lines, transformer stations, and grid enhancement, it also increasingly requires investment in supporting technologies and operational capabilities to deal with more active networks (e.g. bi-directional flows) and more volatility. Alliander has established a separate System Operations unit \u0026ndash; like in transmission \u0026ndash; given the increased active nature of their grid. The regulator has also recognized the impact of the energy transition and is considering moving to a more forward-looking input-based system to account for increasing investments related to the energy transition (ACM \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThere have been several studies examining the impact of the energy transition on network costs.\u003csup\u003e16\u003c/sup\u003e In a study by Montfoort \u003cem\u003eet al.\u003c/em\u003e (2024) using data from all the Dutch DSOs from 2012\u0026ndash;2020, the authors find that energy transition variables, such as solar PV, wind onshore, and EV charging points, have a significant and substantial impact on unit costs.\u003csup\u003e17\u003c/sup\u003e Wangsness \u0026amp; Halse (\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) using a panel of more than 100 Norwegian DSOs between 2008\u0026ndash;2017 estimate the effect of growth in EVs on distribution network costs. They find that increasing penetration of EVs increases network costs. Just \u0026amp; Wetzel (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) find that distributed generation is a significant cost driver for German DSOs. Their results indicate that a 10 percent increase in distributed generation capacity leads to a total cost increase of about 1.6 to 2.3 percent \u0026ndash; although not necessarily on a per unit basis.\u003c/p\u003e\u003cp\u003eThere are also technical and theoretical studies into the impact of EVs. Gupta et al. (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), find that an EV penetration rate of 33 percent results in significant upgrades to transformer stations and line reinforcements. Fernandez et al. (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) estimate that integration of EVs requires additional investments in distribution networks up to 20 percent. Gust et al. (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) apply a theoretical model for designing distribution networks with significant demand coincidence (e.g. from EVs and heat pumps) to a large sample of Swiss DSOs. They find that high demand coincidence (e.g. by charging at the same time) increases average network cost by 84 percent (159 percent in the worst-case scenario where the coincidence factor is 1). Additionally, the impact of demand coincidence is greater on large networks and networks with connection density.\u003c/p\u003e\u003cp\u003eHowever, analysis by Lundgren \u0026amp; Vesterberg (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) of 179 Swedish DSOs between 2014 and 2021 shows that increasing small-scale generation and an increasing number of EVs do not significantly affect the technical efficiency of DSOs.\u003c/p\u003e\u003cp\u003eAlthough there has been strong growth in the installed capacity of distributed generation and EV charge points, Dutch electricity demand has remained relatively flat. Our data shows that the average transported volume per connection has remained relatively constant between 2015 and 2024 (circa 10MWh per connection). We also find that network length per connection has remained constant (circa 35m per connection), suggesting that the grid has been expanding in capacity rather than utilisation. Overall Dutch electricity demand increased between 2000 and 2008, but has been relatively flat until the COVID-19 crisis. Dividing national demand for electricity by the number of connections, shows that demand has been declining since 2010 (from 15MWh in 2010 to 12.5MWh in 2024). Part of this reduction in demand is the result of energy efficiency, deindustrialisation, and the shifting of production \u0026ndash; via solar PV \u0026ndash; to behind-the-meter.\u003c/p\u003e\u003cp\u003eThe quality of the distribution network has remained stable, even with cost pressure from lower tariffs, increased investments, and the increase in intermittency. The average SAIDI has fluctuated between 20 and 30 minutes\u003csup\u003e18\u003c/sup\u003e, with an average of 22.2 minutes between 2004 and 2024. This SAIDI is extremely low when compared internationally. In Europe the average reported SAIDI is around 174 minutes, based on a selected group of DSOs for each capital city.\u003csup\u003e19\u003c/sup\u003e\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cem\u003eProductivity growth over time\u003c/em\u003e\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cp\u003eTo assess productivity growth, we compare the change in output (number of connections) with the change in input (operational costs and capital costs, excluding transmission costs and network losses). This is a gross output or value-add approach, rather than the more classic Total Factor Productivity (TFP). Ideally, we would like to examine the utilisation of the network over time, and not just the number of connections, given that load factors and capacity utilisation also drive costs.\u003c/p\u003e\u003cp\u003e[\u003cem\u003eFigure 6 - Productivity growth indexed (2000\u0026thinsp;=\u0026thinsp;100)\u003c/em\u003e]\u003c/p\u003e\u003cp\u003eThe average productivity growth of the sector was 0.3 percent per year between 2000 and 2024. Over this period annual average operational cost productivity growth was \u0026minus;\u0026thinsp;0.8 percent and annual average capital cost productivity growth was 1.2 percent. This is the result of an increase in operational costs and increase in the allowed return driven by higher investment levels, whilst the growth in the number of connections remains constant (1.1 percent per year). The overall increase in productivity is in large part the result of the decrease in the allowed rate of return (see Fig.\u0026nbsp;2 and Fig.\u0026nbsp;3). This WACC-effect allows the capital productivity to increase.\u003c/p\u003e\u003cp\u003eAs discussed earlier, from 2018 the investment levels of the sector started to increase (see Fig.\u0026nbsp;4) because of the energy transition. Comparing the period before (2000\u0026ndash;2017) and after the acceleration of the energy transition (2018\u0026ndash;2024), shows a significant difference in productivity growth. Overall average productivity growth of the sector was 1.9 percent (2000\u0026ndash;2017) compared to -3.6 percent (2018\u0026ndash;2024). Operational cost productivity growth per year was 1.2 percent between 2000-17 but fell to -5.5 percent between 2018-24. As for the capital cost productivity, this dropped from 2.3 percent per year before the energy transition acceleration in 2018 to -1.4 percent between 2018-24.\u003c/p\u003e\u003cp\u003eThe sector compares favourably with the TFP of eight sectors across 11 European countries (see ACM \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) over the full period, which finds an average productivity growth of 0.5 percent between 1995 and 2017. The developments up to the energy transition inflection point are also positive compared to other studies. Just \u0026amp; Wetzel (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) analyse the cost efficiency of German DSOs between 2011 and 2017, using a dataset with 450 DSOs. They find an average cost reduction potential in the range of 12\u0026ndash;18 percent taking both transient and persistent productivity factors into account, which is what we find for the same period on a per connection basis. Amundsveen \u0026amp; Kvile (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) find average annual productivity increase of 1.3 percent between 2004 and 2014 for Norwegian DSOs \u0026ndash; across three regulatory periods. In our sample we observe a similar increase (1.9 percent). They find evidence of a positive impact on efficiency from \u0026ldquo;smarter\u0026rdquo; networks. In addition, larger DSOs have higher average productivity \u0026ndash; possibly related to scale economies with deploying smart technologies in networks.\u003c/p\u003e\u003cp\u003eAjayi et al. (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) analyse the productivity growth of distribution networks in Great Britain and how changes in incentive mechanisms have influenced productivity. Over the 1990/91\u0026ndash;2018/19 period the find productivity growth of approximately 1 percent per year. Ramos-Real et al. (\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2009\u003c/span\u003e) using a panel of 18 distribution companies in Brazil between 1998 and 2005 find annual productivity growth of 1.3 percent. The productivity of the Australian electricity distribution sector, on the other hand, decreased over 2006-15 period at an average yearly rate of 1.4 percent (AER, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Following this, it trended up over the period 2015\u0026ndash;23 with 0.9 percent per year. The improvement is the result of lower operating costs.\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cem\u003eImpact on costs and performance by regulatory period\u003c/em\u003e\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cp\u003eAlthough the Constructed Regulated Revenue is not perfectly comparable on a year-by-year basis given missing data and necessary estimations, the underlying changes in reporting, and the timing effects of regulation, we nevertheless analyse the impact of the regulatory periods on the financial performance of the sector. The results should be treated with caution and provide directional information, rather than an accurate impact assessment. Table\u0026nbsp;4 shows the changes in the different revenue components between the regulatory periods.\u003c/p\u003e\u003cp\u003e[\u003cem\u003eTable\u0026nbsp;4 \u0026ndash; Change in underlying components of the Constructed Regulated Revenue per Regulatory Period\u003c/em\u003e]\u003c/p\u003e\u003cp\u003eAs discussed earlier, Table\u0026nbsp;4 shows the impact of the lower WACC on the composition of the revenue per connection. Whereas in the first regulatory period, the return component made up 31 percent of the average cost per connection, in 2024 that has dropped to 18 percent. The share of operational costs has conversely increased in importance, from 38 percent in 2000 to 48 percent in 2024. The depreciation component has been stable over all the periods. Losses remained relatively stable, except the sixth regulatory period when DSOs benefited from low wholesale electricity prices. The average network loss compensation in the current regulatory period are similar to at the start of regulation. In the third and sixth regulatory periods there is the biggest downward correction in allowed return in absolute terms, circa \u0026euro;15 and \u0026euro;25 per connection respectively. In the seventh regulatory period operational costs per connection show a big increase \u0026ndash; \u0026euro;10 per connection \u0026ndash; following the increase in investments.\u003c/p\u003e\u003cp\u003eFigure 7 presents the indexed development of the DSO sector revenue using the X-factors set by the ACM and compare this with the development of our Constructed Regulated Revenue.\u003csup\u003e20\u003c/sup\u003e This gives a view on how accurately projected revenues, based on X-factors, track actual costs, and whether there is over or under compensation.\u003c/p\u003e\u003cp\u003e[\u003cem\u003eFigure 7 - Constructed Regulated Revenue vs. ACM Projected Revenue (index 2000\u0026thinsp;=\u0026thinsp;100)\u003c/em\u003e]\u003c/p\u003e\u003cp\u003eIn the first regulatory period the sector reduced costs quicker than the X-factor target, generating a cost-efficiency gain of \u0026euro;105mln relative to the projected revenue. Sector costs stabilise in the second regulatory period, whilst allowed revenues decrease due to the 2.7 percent X-factor. The shortfall in revenue in the second period equals \u0026euro;296mln. In the third regulatory period there is a limited correction, with a negative X-factor of 0.7 percent.\u003csup\u003e21\u003c/sup\u003e This results in a gain of \u0026euro;62mln for the sector relative to the projected revenue (2000 figures). The sum of the first three regulatory periods is a total revenue shortfall of \u0026euro;130mln. That is, sector costs were higher than allowed revenues. This shortfall is equal to approximately 1 percent of total sector costs over the three regulatory periods. Given some of the data issues we noted earlier, it seems plausible to assume that revenues matched costs across the first three regulatory periods (2001\u0026ndash;2007).\u003c/p\u003e\u003cp\u003eIn the fourth regulatory period (2008\u0026ndash;2010), the X-factor increases significantly to 5.2 percent. However, sector costs rise during this period, leading to \u0026euro;530mln shortfall between the allowed revenues and the actual underlying sector costs. To correct for this, the X-factor in the fifth regulatory period becomes negative (essentially allowing tariffs to increase). This leads to a surplus of \u0026euro;393mln for the sector. The reset of the starting cost base in the sixth regulatory period provides a higher starting point, but with a positive X-factor of 4.4 percent, allowed revenues decline again. The combined effect of these three regulatory periods is a \u0026euro;60mln shortfall \u0026ndash; equivalent to 0.3 percent of total sector costs over the three regulatory periods. Based on this outcome, we also assume that allowed revenues matched sector costs between 2008\u0026ndash;2016.\u003c/p\u003e\u003cp\u003eThe start of the seventh regulatory period in 2017 coincides with the inflection point in the underlying investment trend due to the energy transition. It also marks the first five-year regulatory period \u0026ndash; in theory allowing for a more long-term and stable tariff income outlook. In addition, the real pre-tax WACC is further lowered from 4.3 percent at the start of the period to 2.8 percent at the end of the period. Although allowed revenues decline, sectoral costs do not. This results in a shortfall of approximately \u0026euro;1bn (~\u0026thinsp;10 percent of total sector costs over the 5-year period). The X-factor for the eighth, and current, period takes this into account and is negative (-3.2 percent). However, the accelerating investment agenda, higher operational costs, and increased loss expenses, increase \u0026ndash; rather than decrease \u0026ndash; the discrepancy between underlying costs and allowed revenues. Although the regulatory period ends in 2026, the current shortfall is already more than \u0026euro;1.5bn \u0026ndash; circa 20 percent of the total costs for the current three years of the regulatory period.\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;5 summarises the findings per regulatory period and compares this with some of the underlying drivers of the change in costs.\u003c/p\u003e\u003cp\u003e[\u003cem\u003eTable\u0026nbsp;5 \u0026ndash; Change in Constructed Regulated Revenue per Connection per Regulatory Period\u003c/em\u003e]\u003c/p\u003e\u003cp\u003eAverage installed renewable capacity triples from the sixth to the seventh period and from the seventh to the currently still ongoing eight regulatory period. EV charging capacity shows a similar pattern. As for heat pumps, this is still at the start of the s-curve. As previously discussed, investment levels also start to increase. Whereas in the sixth regulatory period the fixed asset base grew by \u0026euro;406mln, in the seventh period it increased by \u0026euro;3.6bn, and in the current regulatory period it has so far grown further by \u0026euro;5.1bn, with two years remaining. In a backward-looking regulatory regime, it is difficult to observe underlying changes in a timely manner \u0026ndash; and it requires judgement whether an observed change is transient or structural, and in the case of the investments in the energy transition in the Netherlands, whether they are accelerating. Yardstick competition has been well documented in prior literature when sectoral costs are stable, and changes occur gradually, and the magnitude is limited. It is less effective in situations where costs and revenues are more volatile, large structural shifts occur in short time frames, and when the pace also differs between DSOs (ACM \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2012\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThere has been a lively discussion whether the regulator should have altered its approach and whether there was sufficient evidence that investments would be rising significantly, and that ex-ante yardstick competition was not providing incentives for DSOs to invest ahead of demand (Hensgens et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2021\u003c/span\u003e, Nieuwsuur, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e\u003cem\u003eBenefits of yardstick regulation\u003c/em\u003e\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cp\u003eTo assess the overall benefit of yardstick regulation, we compare the Constructed Regulated Revenue with a counterfactual revenue in absence of regulation. We assume that network losses, fixed assets, and depreciation are the same in the counterfactual situation. Operational costs per connection are assumed equal to 2000 and scaled with the number of connections in each year.\u003csup\u003e22\u003c/sup\u003e For the allowed return we follow McKinsey (2002) and assume that the cost of equity remains stable over time.\u003csup\u003e23\u003c/sup\u003e We use the average of the low and high estimate of the cost of equity estimates from the ACM for the first and second regulatory period as the basis for the full period (6.0-9.1 percent). We then apply the estimated cost of debt by the ACM using a 60 percent leverage ratio. We then discount the difference between our Constructed Regulated Revenue and the counterfactual using the prescribed societal cost-benefit discount rate to assess government policy changes (2.25 percent real) (see PBL, \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eBased on our analysis we estimate the total benefits, in Net Present Value (NPV) terms, from yardstick regulation for the period 2000 to 2024 to be equal to \u0026euro;2.6bn or equal to \u0026euro;294 per connection (both in real terms 2000). Interpreted as a perpetual annuity, the estimated NPV of \u0026euro;294 corresponds to approximately 2.5 percent of the Constructed Regulated Revenue per connection at the start of regulation. This implies that, on average, the regulatory effect over the period 2000\u0026ndash;2024 is equivalent to a permanent efficiency gain of about 2.5 percent of yearly revenues per connection.\u003c/p\u003e\u003cp\u003eOur findings are significantly lower than what other studies have found. Haffner \u0026amp; Meulmeester (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2005\u003c/span\u003e) estimate the total cumulative benefit of regulation at \u0026euro;1.1bn for the 2001\u0026ndash;2006 period. This is calculated by estimating the value of the tariff reductions without considering the counterfactual or taking the net present value. They also find that in real terms the energy bill has decreased by 12 percent over this period. In our analysis we find a decrease of 7 percent. Mulder \u0026amp; Plug (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2009\u003c/span\u003e) calculate that in 2010 the cumulative benefits of regulation were equivalent to \u0026euro;6bn. Here also there is no clear counterfactual, no discounting, and it seems that the authors assume that current tariffs would remain in place in absence of regulation.\u003c/p\u003e"},{"header":"5. Discussion and Policy Implications","content":"\u003cp\u003eThe evolution of the Dutch electricity distribution sector over the past quarter-century offers a compelling case study of how economic regulation, technological shifts, and policy transitions interact to shape network performance, costs, and productivity. This analysis, spanning 2000 to 2024, reveals a sector that initially achieved substantial efficiency gains and productivity growth under a robust regulatory framework but now faces growing strain from external shocks, accelerated investment needs, and regulatory lag. The story is not one of linear progress, but of a sector and regulator adapting\u0026mdash;at times reactively\u0026mdash;to a rapidly transforming energy landscape.\u003c/p\u003e\u003cp\u003eAt the start of regulation in 2000, the constructed regulated revenue per connection stood at \u0026euro;284 (in 2000 prices). Over 24 years, this increased to \u0026euro;381, reflecting a modest 1.2 percent annual real growth. However, this aggregate figure conceals substantial internal variation. From 2000 to 2022, network costs declined by nearly 15 percent in real terms, due in large part to declining cost of capital, and growing capital efficiency. It is only from 2021 onward that this trend reversed.\u003c/p\u003e\u003cp\u003eThis reversal is the result of the underlying acceleration in the energy transition, with significant growth in renewable capacity from solar PV and wind onshore and significant growth in electrification of transport and heat, which started around 2017, leading to a significant year-on-year increase in capital investments. The investments by the DSOs were then amplified by investments in the transmission network (transmission costs tripled in three years). And although these costs are passed-through, they impact end-user bills. The energy crisis in 2022 further (temporarily) pushed up costs, such as compensation for network losses. This dichotomy over the full period is also clear when looking at the productivity of the sector. Overall productivity growth was 0.3 percent per year. However, between 2000\u0026ndash;2017 it was 1.9 percent and between 2018\u0026ndash;2024 it dropped to -3.6 percent.\u003c/p\u003e\u003cp\u003eOne of the most notable drivers of lower revenues per connection has been the consistent decline in the allowed return on capital. The regulatory real pre-tax WACC fell from 6.7 percent to 3.5 percent, mirroring broader macroeconomic trends in interest rates. This translated into a 49 percent reduction in the allowed return per connection\u0026mdash;from \u0026euro;83 to \u0026euro;43. Early in the regulatory period, capital intensity was also decreasing, fixed assets per connection fell by 25 percent from 2000 to 2017, reflecting improved capital efficiency. These developments combined to significantly reduce capital costs, particularly in the first three regulatory periods. From 2018 onward, however, this picture began to shift as investments started to increase and drove an upswing in capital intensity. Fixed assets per connection increased by 31 percent between 2018 and 2024, returning to 2000 levels.\u003c/p\u003e\u003cp\u003eOperational costs followed a similar but lagged trajectory. They declined steadily by over 21 percent between 2000 and 2018 but have risen since. By 2024, operational costs per connection were 14.3 percent above their 2000 level, reversing two decades of efficiency gains. This increase is partly structural, linked to the operational complexity of integrating variable renewables, managing bi-directional power flows, and accommodating demand-side technologies such as heat pumps and EV chargers. DSOs are no longer simply maintaining a passive infrastructure\u0026mdash;they are operating active, intelligent networks.\u003c/p\u003e\u003cp\u003eThe regulatory framework\u0026mdash;anchored in yardstick competition and the X-factor approach\u0026mdash;has generally delivered cost discipline and incentives for efficiency. However, its backward-looking design has proven inadequate during periods of sudden structural change. The analysis of regulatory periods shows that while surpluses and shortfalls largely balanced out in the early phases, the last two regulatory periods saw substantial under-compensation: a \u0026euro;1\u0026nbsp;billion shortfall in the seventh period (2022\u0026ndash;2026) and a further \u0026euro;1.5\u0026nbsp;billion shortfall to date in the eighth period. These shortfalls are equivalent to 10\u0026ndash;20 percent of sector costs in the respective periods (in previous periods the shortfall was less than 1 percent). They arise from misalignment between allowed revenues and actual cost developments, particularly in the face of surging investment needs and volatile pass-through expenses.\u003c/p\u003e\u003cp\u003eThis misalignment is not merely a financial concern for DSOs\u0026mdash;it has real implications for the energy transition. If DSOs are under-incentivized to invest ahead of demand, or if regulatory lag deters capital deployment, the pace and cost of decarbonization may be adversely affected. While yardstick regulation excels in steady-state environments with gradual change, it is less suited to contexts requiring anticipatory investment and network transformation. Indeed, the evidence of catch-up investment by DSOs post-2020 suggests that earlier regulatory assumptions underestimated the scale and speed of the transition.\u003c/p\u003e\u003cp\u003eThe analysis ultimately suggests that the regulatory model that served the sector and end-users well for much of the early 2000s must now evolve to meet new challenges. Yardstick regulation has delivered cost savings equal to 2.5 percent per year on a per-connection basis. However, tariffs have now started to rise to accommodate rising investments. The ACM has announced that it intends to move to a more forward-looking input-based model, capable of aligning revenue allowances with expected investments in grid flexibility, digitalization, and electrification. Stability and predictability in regulatory incentives are especially important in a sector that is simultaneously managing legacy infrastructure and preparing for a more dynamic, decentralized energy system. The question is whether with better anticipation the reversal in productivity could have been blunted and whether an earlier switch to a more forward-looking regulatory regime would have delivered higher outputs with less costs.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eP.N. wrote the main manuscript text and prepared the figures and tables. All authors reviewed the manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eACM factsheets, multiple years.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eACM X-factorbesluiten RNB Elektriciteit, multiple regulatory periods.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eACM (2012). \u003cem\u003eDe Toekomst van Tariefregulering\u003c/em\u003e, Report prepared for the ACM by PwC, 7 September 2012.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eACM (2015). \u003cem\u003eRapportage winsten regionale netbeheerders elektriciteit en gas 2008\u0026ndash;2013\u003c/em\u003e, 31 maart 2015.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eACM (2020). \u003cem\u003eFrontier Shift for Dutch Gas and Electricity TSOs\u003c/em\u003e, Report prepared for the ACM by Economic Insights, 1 May 2020.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eACM (2023). \u003cem\u003eToekomstige reguleringsmethode netbeheerders voor de reguleringsperiode vanaf 2027\u003c/em\u003e, ACM/22/176591 / Documentnr. ACM/UIT/605442.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAER (2024). \u003cem\u003eAnnual Benchmarking Report, Electricity Distribution Network Service Providers\u003c/em\u003e, November 2024.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAjayi, V., Dolphin, G., Anaya, K., \u0026amp; Pollitt, M. G. (2020). The Productivity Puzzle in Network Industries: Evidence from the Energy Sector, \u003cem\u003eWorking Papers EPRG 2021\u003c/em\u003e, Energy Policy Research Group, Cambridge Judge Business School, University of Cambridge.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAjayi, V., Anaya, K., \u0026amp; Pollitt, M. G. (2022). Incentive regulation, productivity growth and environmental effects: the case of electricity networks in Great Britain, \u003cem\u003eEnergy Economics\u003c/em\u003e, Volume 115. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.eneco.2022.106354\u003c/span\u003e\u003cspan address=\"10.1016/j.eneco.2022.106354\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAlliander (2009\u0026ndash;2024), Annual Reports.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAmundsveen, R., \u0026amp; Kvile, H. M. (2017). Productivity Development Among Norwegian Electricity Distribution System Operators - The Impact of Capital Assessment and Technological Progress. \u003cem\u003eData Envelopment Analysis Journal\u003c/em\u003e, \u003cem\u003e3\u003c/em\u003e(1\u0026ndash;2), 93\u0026ndash;118. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1561/103.00000016\u003c/span\u003e\u003cspan address=\"10.1561/103.00000016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBaarsma, B., de Nooij, M., Koster, W., \u0026amp; van der Weijden, C. (2007). Divide and rule. The economic and legal implications of the proposed ownership unbundling of distribution and supply companies in the Dutch electricity sector. \u003cem\u003eEnergy Policy\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(3), 1785\u0026ndash;1794. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.enpol.2006.05.002\u003c/span\u003e\u003cspan address=\"10.1016/j.enpol.2006.05.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBaumol, W. J. (1982). Productivity Incentive Clauses and Rate Adjustment for Inflation, \u003cem\u003ePublic Utilities Fortnightly\u003c/em\u003e, 110, July, pp. 11\u0026ndash;18.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBruegel (2025). \u003cem\u003eUpgrading Europe\u0026rsquo;s Electricity Grid is About More than Just Money\u003c/em\u003e, Bruegel Policy Brief 04/25, February 2025.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCPB (2000). \u003cem\u003eYardstick competition: Theory, design, and practice\u003c/em\u003e, CPB Netherlands Bureau for Economic Policy Analysis, Working Paper, No. 133, December 2000, The Hague.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCPB (2023). National Productivity Board 2022 Annual Report.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDTe (1999). \u003cem\u003ePrice Cap Regulation in the Electricity Sector, Information and Consultation Document\u003c/em\u003e, July 1999, The Hague.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDuma, D., Pollitt, M. G., \u0026amp; Covatariu, A. (2024). Regulatory Learning in the face of Net Zero Climate Policy: The Case of the UK. \u003cem\u003eReview of Industrial Organization\u003c/em\u003e, \u003cem\u003e65\u003c/em\u003e(2), 505\u0026ndash;530. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11151-024-09972-y\u003c/span\u003e\u003cspan address=\"10.1007/s11151-024-09972-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEneco Holding (2013\u0026ndash;2016), Annual Reports.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEnexis (2015\u0026ndash;2017), Reguleringsdata.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEnexis Holding (2009\u0026ndash;2024), Annual Reports.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEnexis Netbeheer (2013\u0026ndash;2024), Aanvulling segmentinformatie.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEnexis Netbeheer (2009\u0026ndash;2024), Annual Reports.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEuropean Heat Pump Association (2025). \u003cem\u003eHeat pump sales 14 times greater in lead countries\u003c/em\u003e, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ehpa.org/news-and-resources/news/heat-pump-sales-14-times-greater-in-lead-countries/\u003c/span\u003e\u003cspan address=\"https://www.ehpa.org/news-and-resources/news/heat-pump-sales-14-times-greater-in-lead-countries/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEyre, S., \u0026amp; Pollitt, M. G. (2016). Introduction. In S. Eyre, \u0026amp; M. G. Pollitt (Eds.), \u003cem\u003eCompetition and Regulation in Electricity Markets\u003c/em\u003e (pp. xiii\u0026ndash;xxxv). Edward Elgar.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFernandez, L. P., San Roman, T. G., Cossent, R., Domingo, C. M., \u0026amp; Frias, P. (2010). Assessment of the impact of plug-in electric vehicles on distribution networks. \u003cem\u003eIEEE Transactions on Power Systems\u003c/em\u003e, \u003cem\u003e26\u003c/em\u003e(1), 206\u0026ndash;213. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/TPWRS.2010.2049133\u003c/span\u003e\u003cspan address=\"10.1109/TPWRS.2010.2049133\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFilippini, M., \u0026amp; S\u0026aacute;nchez, L. O. (2014). Applications of the stochastic frontier approach in the analysis of energy issues. \u003cem\u003eEconomics and Business Letters\u003c/em\u003e, \u003cem\u003e3\u003c/em\u003e(1), 35\u0026ndash;42. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.17811/ebl.3.1.2014.35-42\u003c/span\u003e\u003cspan address=\"10.17811/ebl.3.1.2014.35-42\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFinancial Times (2025). \u003cem\u003eNetherlands Rations Electricity to Ease Power Grid Stresses\u003c/em\u003e, 13 July 2025.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGupta, R., Pena-Bello, A., Streicher, K. N., Roduner, C., Farhat, Y., Thoeni, D., Patel, M. K., \u0026amp; Parra, D. (2021). Spatial analysis of distribution grid capacity and costs to enable massive deployment of pv, electric mobility and electric heating, \u003cem\u003eApplied Energy\u003c/em\u003e, 287(C). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.apenergy.2021.116504\u003c/span\u003e\u003cspan address=\"10.1016/j.apenergy.2021.116504\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGust, G., Schl\u0026uuml;ter, A., Feuerriegel, S., \u0026Uacute;beda, I., Lee, J. T., \u0026amp; Neumann, D. (2024). Designing Electricity Distribution Networks: The Impact of Demand Coincidence. \u003cem\u003eEuropean Journal of Operational Research\u003c/em\u003e, \u003cem\u003e315\u003c/em\u003e(1), 271\u0026ndash;288. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ejor.2023.11.029\u003c/span\u003e\u003cspan address=\"10.1016/j.ejor.2023.11.029\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHaffner, R. C. G., \u0026amp; Meulmeester, P. (2005). Evaluatie van de Regulering van het Elektriciteitsnetwerk. \u003cem\u003eEconomische en Statistische Berichten\u003c/em\u003e, \u003cem\u003e4472\u003c/em\u003e, 430\u0026ndash;433.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHensgens, R., Nillesen, P. H. L., en, \u0026amp; Wessels, R. (2021). Energietransitie Vergt Aanpassing Regulering Netbeheerders. \u003cem\u003eEconomische en Statistische Berichten\u003c/em\u003e, \u003cem\u003e106\u003c/em\u003e(4804), 578\u0026ndash;581.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHinz, F., Schmidt, M., \u0026amp; Moest, D. (2018). Regional distribution effects of different electricity network tariff designs with a distributed generation structure: The case of Germany. \u003cem\u003eEnergy Policy\u003c/em\u003e, \u003cem\u003e113\u003c/em\u003e, 97\u0026ndash;111. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.enpol.2017.10.055\u003c/span\u003e\u003cspan address=\"10.1016/j.enpol.2017.10.055\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHoefer, T., \u0026amp; Madlener, R. (2021). Locational (In) efficiency of renewable energy feed-in into the electricity grid: A spatial regression analysis. \u003cem\u003eThe Energy Journal\u003c/em\u003e, \u003cem\u003e42\u003c/em\u003e(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5547/01956574.42.1.thof\u003c/span\u003e\u003cspan address=\"10.5547/01956574.42.1.thof\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJamasb, T., \u0026amp; Pollitt, M. G. (2001). Benchmarking and regulation: International electricity experience. \u003cem\u003eUtilities Policy\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(3). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0957-1787(01)00010-8\u003c/span\u003e\u003cspan address=\"10.1016/S0957-1787(01)00010-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. pp. 107\u0026thinsp;\u0026ndash;\u0026thinsp;30.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJoskow, P. L. (2014). Incentive Regulation in Theory and Practice: Electricity Distribution and Transmission Networks, in Rose, N.L. (Ed.), \u003cem\u003eEconomic Regulation and Its Reform: What Have We Learned?\u003c/em\u003e NBER.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJoskow, P. L., \u0026amp; Schmalensee, R. (2025). Cost-of-service regulation of electricity distribution services in the US, in Glachant, J-M., Joskow, P.L., \u0026amp; Pollitt, M.G. (Eds.), \u003cem\u003eHandbook on Electricity Regulation\u003c/em\u003e, Edward Elgar.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJust, L., \u0026amp; Wetzel, H. (2020). Distributed Generation and Cost Efficiency of German Electricity Distribution Network Operators, \u003cem\u003eEWI Working Paper\u003c/em\u003e, No. 20/09.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKendrick, J. W. (1975). Efficiency Incentives and Cost Factors in Public Utility Automatic Revenue Adjustment Clauses. \u003cem\u003eBell Journal of Economics\u003c/em\u003e, \u003cem\u003e6\u003c/em\u003e(1), 299\u0026ndash;313.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiander (2008). 2011\u0026ndash;2024), Annual Reports.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLittlechild, S. (1983). \u003cem\u003eRegulation of British Telecommunications\u0026rsquo; Profitability\u003c/em\u003e, Report to The Secretary of State, Department of Industry, February 1983.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLundgren, T., \u0026amp; Vesterberg, M. (2024). Efficiency in Electricity Distribution in Sweden and the Effects of Small-scale Generation, Electric vehicles and Dynamic tariffs. \u003cem\u003eJournal of Productivity Analysis\u003c/em\u003e, \u003cem\u003e62\u003c/em\u003e, 121\u0026ndash;137. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11123-024-00724-4\u003c/span\u003e\u003cspan address=\"10.1007/s11123-024-00724-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMcKinsey (2002). \u003cem\u003eThe Real Cost of Equity, Perspectives on Corporate Finance and Strategy\u003c/em\u003e, McKinsey on Finance, Number 5, Autumn 2002.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMinistry of Economic Affairs (2003). \u003cem\u003eWet van 2 juli 1998, houdende regels met betrekking tot de productie, het transport en de levering van elektriciteit (Elektriciteitswet 1998)\u003c/em\u003e, (Stb. 2003, 235), The Hague.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMontfoort van, F., Dijkstra, P. T., \u0026amp; Mulder, M. (2024). The impact of energy transition on distribution network costs and effectiveness of yardstick competition: an empirical analysis for the Netherlands. \u003cem\u003eJournal of Regulatory Economics\u003c/em\u003e, \u003cem\u003e65\u003c/em\u003e, 85\u0026ndash;107. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11149-024-09471-8\u003c/span\u003e\u003cspan address=\"10.1007/s11149-024-09471-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMulder, M., \u0026amp; Plug, P. (2009). Reguleren van energiemarkten: sturen in een dynamische omgeving in De blijvende uitdaging: \u003cem\u003eLiber amicorum voor ir. Gert J.L. Zijl en drs. Ren\u0026eacute; Jansen\u003c/em\u003e, NMa Netherlands Competition Authority.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNetbeheer Nederland (2024). \u003cem\u003eFinanci\u0026euml;le Impact Energietransitie voor Netbeheerders (FIEN+)\u003c/em\u003e, PwC, 16 December 2024.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNiesten, E. (2010). Network investments and the integration of distributed generation: Regulatory recommendations for the Dutch electricity industry. \u003cem\u003eEnergy Policy\u003c/em\u003e, \u003cem\u003e38\u003c/em\u003e(8), 4355\u0026ndash;4362.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNieuwenburg, M., Smit, E., \u0026amp; Mulder, M. (2020). Performance of Dutch energy distribution operators Performance of Dutch energy distribution operators an assessment of tariffs, quality and financial position since 2010 an assessment of tariffs, quality and financial position since 2010. \u003cem\u003eCentre for Energy Economics Research (CEER)\u003c/em\u003e, Policy Papers 7.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNieuwsuur (2022). \u003cem\u003eReconstructie: Hoe Toezichthouder Tekorten op het Elektriciteitsnet Mede Veroorzaakte\u003c/em\u003e, maandag 3 oktober 2022 16.00.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://nos.nl/nieuwsuur/artikel/2446934\u003c/span\u003e\u003cspan address=\"https://nos.nl/nieuwsuur/artikel/2446934\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNillesen, P. H. L., \u0026amp; Pollitt, M. G. (2007). The 2001-3 electricity distribution price control review in the Netherlands: regulatory process and consumer welfare. \u003cem\u003eJournal of Regulatory Economics\u003c/em\u003e, \u003cem\u003e31\u003c/em\u003e, 261\u0026ndash;287. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11149-006-9014-8\u003c/span\u003e\u003cspan address=\"10.1007/s11149-006-9014-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNillesen, P. H. L., \u0026amp; Pollitt, M. G. (2021). Ownership Unbundling of Electricity Distribution Networks. \u003cem\u003eEconomics of Energy and Environmental Policy\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(1), 147\u0026ndash;158. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5547/2160-5890.10.1.pnil\u003c/span\u003e\u003cspan address=\"10.5547/2160-5890.10.1.pnil\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNMa. (2011). \u003cem\u003eOnderzoek naar de methodologie voor de verdeling van de kosten van netverliezen\u003c/em\u003e (Vol. 29). KEMA and SEO. March 2011.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePBL (2022). \u003cem\u003eMaatschappelijke Kosten-Batenanalyse en Brede Welvaart; Een Aanvulling op de Algemene MKBA-Leidraad\u003c/em\u003e, June 2020.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePBL (2024). \u003cem\u003eLaadinfrastructuur mobiliteit, 2018\u0026ndash;2023\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ehttps://\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003emonovi.pbl.nl/indicatoren/nl059902-laadinfrastructuur-mobiliteit-2018-2023#bronnen\u003c/span\u003e\u003cspan address=\"http://monovi.pbl.nl/indicatoren/nl059902-laadinfrastructuur-mobiliteit-2018-2023#bronnen\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePollitt, M. G., Duma, D., Covatariu, A., \u0026amp; Nillesen, P. H. L. (2025). A Global Map of Electricity and Gas Distribution Network Companies, \u003cem\u003eEnergy Policy Research Group, Working Papers\u003c/em\u003e No.2519, Cambridge Judge Business School, University of Cambridge.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRamos-Real, F. J., Tovar, B., Iootty, M., de Almeida, E. F., \u0026amp; PintoJr., H. Q. (2009). The Evolution and Main Determinants of Productivity in Brazilian Electricity Distribution 1998\u0026ndash;2005: An Empirical Analysis, \u003cem\u003eEnergy Economics\u003c/em\u003e, vol. 31(2), pp. 298\u0026ndash;305. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.eneco.2008.11.002\u003c/span\u003e\u003cspan address=\"10.1016/j.eneco.2008.11.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRegulatory Assistance Project (2019). \u003cem\u003ePerformance-based Regulation: Aligning Incentives with Clean Energy Outcomes\u003c/em\u003e, June 2019.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRVO (2025). \u003cem\u003eCharging and Tank Infrastructure in The Netherlands\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ehttps://\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003eduurzamemobiliteit.databank.nl/mosaic/en-us/elektrisch-vervoer/laad--en-tankinfrastructuur-in-nederland\u003c/span\u003e\u003cspan address=\"http://duurzamemobiliteit.databank.nl/mosaic/en-us/elektrisch-vervoer/laad--en-tankinfrastructuur-in-nederland\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchleifer, A. (1985). A theory of yardstick competition. \u003cem\u003eRand Journal of Economics\u003c/em\u003e, 16(3).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSolar Magazine (2025). \u003cem\u003eDe harde cijfers | SDE++: marktgroei 2024 bijgesteld tot bijna 2,2 gigawattpiek\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ehttps://\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003esolarmagazine.nl/nieuws-zonne-energie/i40177/de-harde-cijfers-sde-marktgroei-2024-bijgesteld-tot-bijna-2-2-gigawattpiek\u003c/span\u003e\u003cspan address=\"http://solarmagazine.nl/nieuws-zonne-energie/i40177/de-harde-cijfers-sde-marktgroei-2024-bijgesteld-tot-bijna-2-2-gigawattpiek\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStedin (2016\u0026ndash;2024), Annual Reports.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStedin Groep (2016\u0026ndash;2024), Annual Reports.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTertre (2023). \u003cem\u003eStructural Changes in Energy Markets and Price Implications: Effects of the Recent Energy Crisis and Perspectives of the Green Transition\u003c/em\u003e, European Central Bank.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTweede Kamer der Staten-Generaal. (2005). \u003cem\u003eWijziging van de Elektriciteitswet 1998 en van de Gaswet in verband met nadere regels omtrent een onafhankelijk netbeheer\u003c/em\u003e (Vol. 30212). nr. 3, Memorie van Toelichting.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVlijm, W. (2002). \u003cem\u003eDe Interactie tussen de Overheid en de Elektriciteitssector in Nederland: De Ontwikkeling van het Nutsbedrijf PGEM naar de Energieonderneming Nuon (1916\u0026ndash;2001)\u003c/em\u003e, PhD Dissertation, Radboud Universiteit.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVogelsang, I. (2002). Incentive Regulation and Competition in Public Utility Markets: A 20-Year Perspective. \u003cem\u003eJournal of Regulatory Economics\u003c/em\u003e, \u003cem\u003e22\u003c/em\u003e, 5\u0026ndash;27. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1023/A:1019992018453\u003c/span\u003e\u003cspan address=\"10.1023/A:1019992018453\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWangsness, P. B., \u0026amp; Halse, A. H. (2021). The impact of electric vehicle density on local grid costs: Empirical evidence from Norway. \u003cem\u003eThe Energy Journal\u003c/em\u003e, \u003cem\u003e42\u003c/em\u003e(5). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5547/01956574.42.5.pwan\u003c/span\u003e\u003cspan address=\"10.5547/01956574.42.5.pwan\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Footnotes","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003e Although it was actually first suggested for US telecoms by Baumol (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e1982\u003c/span\u003e) drawing on ideas in Kendrick (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e1975\u003c/span\u003e), as discussed in Eyre and Pollitt (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e For a recent discussion see Joskow and Schmalensee (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e Ministry of Economic Affairs (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2003\u003c/span\u003ea), Wet van 2 juli 1998, houdende regels met betrekking tot\u003c/span\u003e\u003cdiv id=\"Par25\" class=\"Para\"\u003ede productie, het transport en de levering van elektriciteit (Elektriciteitswet 1998) (Stb. 2003, 235), The Hague.\u003c/div\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e In Dutch: Samengestelde Output (SO).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e Following the first Oil Crisis in 1973, Dutch government policy was focused on getting more grip on the energy sector and energy provision. It was active government policy to consolidate the energy sector in the Netherlands, most notably with the report from the \u003cem\u003eCommissie Concentratie Nutsbedrijven (CoCoNut)\u003c/em\u003e in 1980 and the report from the \u003cem\u003eCommissie Brandsma\u003c/em\u003e in 1985. In 1985 there were 158 energy companies. The recommendations were to consolidate this to 56 companies in 1987 and 48 in 1991 (see Vlijm, \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2002\u003c/span\u003e).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e The three smaller DSOs are Coteq (0.6 percent market share), Rendo (0.4 percent market share), and Westland Infra (0.7 percent market share). The three DSOs that were consolidated into three large DSOs were Enduris, NRE, and ONS. There have also been several service area swaps between the three large DSOs.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e IFRS\u0026thinsp;=\u0026thinsp;International Financial Reporting Standards.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e In the first regulatory period 2001\u0026ndash;2003, the ACM determined the starting value for the RAB (start-GAW), which the ACM treats as an investment done at the end of 2000, with one uniform depreciation period.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e We discuss this restructuring in Nillesen and Pollitt (\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e We base the one-off unbundling costs on the submission to Parliament by the Ministry of Economic Affairs in 2005, which provides the financial impact assessment of Deloitte (Tweede Kamer der Staten-Generaal, \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). The costs associated with unbundling (one-off unbundling costs estimated between \u0026euro;80-130mn) were not allowed to be passed through to end-users by law. We assume that all three DSOs incurred the costs in 2009, even though Stedin and Eneco unbundled later following court appeals.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNMa (\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2011\u003c/span\u003e) estimates that losses cost approximately EUR 30 per customer and is equivalent to ~\u0026thinsp;5 percent of total distributed volume (70 percent technical and 30 percent administrative) of which 82 percent occurs at MS/LS level.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e The study projects the total investment in infrastructure for electricity (transmission and distribution), gas (transport and distribution), heat networks, and hydrogen infrastructure to be approximately \u0026euro;200bn up to 2040 (Netbeheer Nederland \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e On a per capita basis, the Netherlands has the highest solar PV penetration in the world.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e There are 614k full electric vehicles and 458k plug-in hybrid vehicles as per July 2025, according to CBS.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e See European Heat Pump Association, 3 July 2025.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e See for example, Hinz et al. (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), Hoefer \u0026amp; Madlener (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), Halse (2021), and Filipini \u0026amp; Sanchez (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e This could have implications for the effectiveness of yardstick composition, as distribution costs increase but the output (volume per connection) remains constant. If different DSOs are exposed to different degrees of energy transition-related investments in the network, the outcome from yardstick competition could lead to over or under compensation for individual DSOs. They observe 0\u0026ndash;10 percent changes in individual price caps for most DSOs, and even an increase of 20 percent for one of the DSOs.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e SAIDI\u0026thinsp;=\u0026thinsp;System Average Interruption Duration Index.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e See Pollitt et al. (\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) where all the DSOs from capitals around the world are compared.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e The effective X-factor for the first regulatory period was 2.26 percent, although the reported X-factor that was negotiated was 3.2 percent. This is the result of the no \u0026ldquo;reformatio in peius\u0026rdquo; principle, which meant that decisions by the regulator, if more favourable, could not be changed.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e The third regulatory period was shortened to one year (instead of three years as initially planned), given the transfer of the 110/150kV lines from the DSOs to TenneT and the change to yardstick regulation (as opposed to the individual benchmarking in the first and second regulatory period). The negotiated outcome for the third period was an overall tariff freeze in reals terms for electricity and gas combined.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e If we assume that unit operational costs develop in line with TFP for the Netherlands, then the benefits of yardstick regulation decline by approximately \u0026euro;500mln to \u0026euro;2.1bn.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e McKinsey (2002) analysis suggests that investors seek a stable total return and that during periods of low interest rates the market risk premium rises and \u003cem\u003evice versa\u003c/em\u003e. This suggests that the equity component of the WACC remains stable over time, rather than vary with the risk-free rate of interest. They estimate the average inflation-adjusted cost of equity implied by stock market valuations each year from 1963 to 2001 in the US and from 1965 to 2001 for the UK to be 7 percent. Our estimate using the average from the ACM is 7.6 percent and in line with the McKinsey estimates.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTables 1 to 5 are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"journal-of-regulatory-economics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"rege","sideBox":"Learn more about [Journal of Regulatory Economics](http://link.springer.com/journal/11149)","snPcode":"11149","submissionUrl":"https://submission.nature.com/new-submission/11149/3","title":"Journal of Regulatory Economics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7601011/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7601011/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis paper assesses the long-term effects of incentive regulation on the financial and operational performance of Dutch electricity distribution system operators (DSOs) from 2000 to 2024. Using a proxy for regulated income—Constructed Regulated Revenue—we evaluate the outcome of yardstick-based benchmarking across seven regulatory periods. Constructed Regulated Revenue per connection rose from €284 in 2000 to €381 in 2024, a real-term increase of 1.2 percent per year. This growth, however, masks substantial heterogeneity: from 2000 to 2017, productivity improved by 1.9 percent per year, but between 2018 to 2024 this dropped to -3.6 percent per year. The backward-looking design of the regulatory framework limited incentives for anticipatory investments. Expenditures rose sharply as DSOs responded to grid congestion and the demands of the energy transition, eroding earlier cost reductions. We conclude that while incentive regulation in the Netherlands was effective in reducing costs during the early periods, it did not adequately support long-term investment under evolving system conditions. The results underline the importance of forward-looking regulatory mechanisms that align cost efficiency and pass-through cost with future grid needs.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eJEL Codes: L94, D24, Q48, L51\u003c/em\u003e\u003c/p\u003e","manuscriptTitle":"Incentive Regulation and Distribution Network Performance: A Dutch Case Study (2000–2024)","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-17 02:23:10","doi":"10.21203/rs.3.rs-7601011/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-09-25T19:12:08+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-17T07:53:42+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-17T07:52:39+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Regulatory Economics","date":"2025-09-12T13:12:36+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-regulatory-economics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"rege","sideBox":"Learn more about [Journal of Regulatory Economics](http://link.springer.com/journal/11149)","snPcode":"11149","submissionUrl":"https://submission.nature.com/new-submission/11149/3","title":"Journal of Regulatory Economics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"27544d01-d51b-460e-b1f8-13750a7750fc","owner":[],"postedDate":"October 17th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-04-27T16:05:33+00:00","versionOfRecord":{"articleIdentity":"rs-7601011","link":"https://doi.org/10.1007/s11149-026-09511-5","journal":{"identity":"journal-of-regulatory-economics","isVorOnly":false,"title":"Journal of Regulatory Economics"},"publishedOn":"2026-04-26 15:58:53","publishedOnDateReadable":"April 26th, 2026"},"versionCreatedAt":"2025-10-17 02:23:10","video":"","vorDoi":"10.1007/s11149-026-09511-5","vorDoiUrl":"https://doi.org/10.1007/s11149-026-09511-5","workflowStages":[]},"version":"v1","identity":"rs-7601011","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7601011","identity":"rs-7601011","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}