Unfunded Mandates and the Economic Impact of Decentralisation. When Finance Does Not Follow Function
Abstract
Decentralisation has frequently been sold as a means to increase well-being and development. Yet, questions remain as to whether decentralisation improves economic performance. This is possibly because decentralisation processes have often led to ‘unfunded mandates’, that is, a mismatch between the powers transferred to subnational tiers of government and the resources allocated to them. In this article, we analyse how unfunded mandates shape regional economic growth across 518 regions in 30 Organisation for Economic Co-operation and Development countries over the period 1997–2018. There is a negative, statistically significant, and robust impact of unfunded mandates on economic growth. This effect is higher in more politically and less fiscally decentralised regions and in regions with a higher level of wealth. Unfunded mandates thus represent a serious drag on the potential positive economic effect of political decentralisation. Hence, for those benefits to materialise, better not more decentralisation – ensuring that finance follows function – should be pursued.
Introduction
Over the past 50 years, the drive towards decentralisation has been almost universal. Today, 40.4% of public expenditure in Organisation for Economic Co-operation and Development countries is undertaken at a subnational level (OECD, 2019).
Understood as the transfer of powers and resources from central to subnational tiers of government (OECD, 2017), decentralisation has been fundamentally sold as a means to deliver greater economic dynamism and well-being for citizens, wherever they live (Rodríguez-Pose and Sandall, 2008). Yet, the verdict on whether decentralisation has delivered economic dividends remains very much in the air (Morgan, 2002). The many theoretical and empirical studies investigating the link between decentralisation and regional economic development have reached inconclusive results. Early theoretical research tended to underline that decentralisation leads to higher economic growth through increased efficiency, policy innovation, and political accountability (Donahue, 1997; Putnam, 1993; Tiebout, 1956). But the empirical evidence remains mixed, with different authors finding different relationships: positive (e.g. Iimi, 2005), negative (e.g. Rodríguez-Pose and Ezcurra, 2011), or statistically insignificant (e.g. Thornton, 2007).
One of the potential reasons for the heterogeneity of results is that past research has mostly focused on estimating the impact of political and fiscal decentralisation on the economy, either separately or jointly. In doing so, these studies make the crucial assumption that ‘finance follows function’: once an intermediate tier of government is awarded new powers, the necessary resources follow (Bahl and Martínez-Vázquez, 2013). Nonetheless, significant gaps between the transfer of political powers and available resources remain, particularly in places where decentralisation has been driven by national authorities, which hold the upper hand in the process. These gaps, referred to as ‘unfunded mandates’, leave subnational governments without adequate funding to fulfil their mandates, a fact that can undermine the purported benefits of decentralisation.
However, the attention unfunded mandates have received in scholarly research has been limited. To our knowledge, no cross-regional empirical study covering a large number of countries exists on the issue of unfunded mandates. Omitting this phenomenon may be one reason behind the heterogeneity of results on the link between decentralisation and economic development.
This study fills this gap by incorporating unfunded mandates into an analysis of the effects of decentralisation on economic growth. Using an original panel dataset of 518 regions in 30 OECD countries for the period 1997–2018, we analyse the extent to which unfunded mandates undermine regional economic growth. We also examine whether and to what extent a region’s political and fiscal decentralisation, as well as its level of development affects the impact of unfunded mandates on regional economic growth. The overall aim of the analysis is thus to go beyond existing knowledge – which fundamentally focuses on the degree of political and fiscal decentralisation – and concentrate on the impact the frequently uneven balance between political and fiscal decentralisation in a particular region may have on its economic development.
The study is structured as follows. In the ‘Decentralisation, Unfunded Mandates, and Economic Growth’ section, we critically review the state of the literature on decentralisation as a development tool and, more specifically, on the issue of unfunded mandates. In the ‘Methodology’ section, we present the research question and hypotheses as well as the methodology and the data. The ‘Results’ section presents the empirical results and their robustness checks. The conclusions are developed in the ‘Conclusion’ section, outlining policy implications and proposing new avenues for future research.
Decentralisation, Unfunded Mandates, and Economic Growth
Decentralisation as a Tool for Development
Theories of decentralisation have generally considered that the transfer of authority and resources to lower tiers of government yields economic benefits. Following Oates’s (1972) fiscal decentralisation theorem, decentralisation should lead to allocative efficiency. Decentralised governments are portrayed as being closer to citizens and more capable of catering for the heterogeneous needs and preferences of their populations (Klugman, 1994). Furthermore, local decision-makers may have a greater incentive than national ones to become more efficient and innovative in the delivery of public goods and services (Donahue, 1997). Otherwise, their constituents may ‘vote with their feet’, moving to other regions offering better services or lower taxes (Musgrave, 1959; Tiebout, 1956). Intensified competition among subnational entities may also trigger productive efficiency as regions specialise in their comparative advantages (Rodríguez-Pose and Bwire, 2004). Finally, the increased proximity of political power to citizens may improve accountability and transparency, increase participation, and reduce corruption (Ebez and Yilmaz, 2002; Putnam, 1993). All these factors put together should result in a more efficient administration and improve economic performance.
The purported benefits of decentralisation can, nevertheless, be challenged. Proximity to citizens per se does not automatically lead to better decision-making. Moreover, as indicated by Prud’homme (1995), policy preferences may not vary significantly among regions. This potential lack of variation in preferences may produce subnational governments that are less – and not more – capable and/or efficient at providing the same goods and services than the central government. Subnational governments may also incur in diseconomies of scale. Transparency and corruption may also not improve through decentralisation, especially when resources are limited. Insufficient human and financial capacity can stymie the capacity to monitor and prevent corruption (De Mello and Barenstein, 2001). More importantly for the purpose of this analysis, Prud’homme considers that decentralised entities are efficient only when they possess sufficient financial and human resources to fulfil their mandates. Hence, the benefits of decentralisation are contingent upon an appropriate funding and staffing of subnational authorities (Rodríguez-Pose and Gill, 2005).
Different empirical studies have tested these assumptions, producing a plurality of results depending on the conceptual, methodological, and econometric approaches adopted (Martínez-Vázquez and McNab, 2003). The results vary between those who find a positive relationship between decentralisation and economic growth (e.g. Iimi, 2005) and those reporting a negative impact (e.g. Rodríguez-Pose and Ezcurra, 2011). In between these extremes, some uncover non-linear, hump-shaped relationships between decentralisation and economic performance (e.g. Bodman, 2008; Thießen, 2003), while others do not find any significant impact of decentralisation on growth (e.g. Feld and Schnellenbach, 2011; Thornton, 2007).
An inconclusive scholarly literature points to the possibility of past analysis of the economic impact of decentralisation omitting relevant variables. The focus so far has fallen squarely on differences in fiscal and political decentralisation, considering them individually. But the fact that the economic impact of decentralisation is bound to be the result of their combination has been overlooked. Yet, if a subnational authority does not receive or garner sufficient financial resources, the degree of fiscal or political decentralisation may be irrelevant for growth in the face of a glaring mismatch between authority and resources. Consequently, the (im)balance between political and fiscal decentralisation – in other words, the presence of unfunded mandates – is probably a more important factor for the capacity of subnational tiers of government to deliver on their economic promise than the degree of political and fiscal decentralisation on their own.
Unfunded Mandates: When Finance Does Not Follow Function
The link between unfunded mandates and economic performance at a regional level has so far received limited attention in decentralisation research. This has, however, not prevented scholars from studying the legal aspects of unfunded mandates and the political incentives that prompt their appearance, especially in the United States (e.g. Adler, 1997; Posner, 1998). An unfunded mandate can be defined as any devolved responsibility not accompanied by the necessary resources to fulfil it (Ross, 2018).1 Briefly stated, it refers to situations where finance does not follow function (Bahl, 1999). This is a controversial matter as central and subnational governments have competing interests. On the one hand, the central government may see unfunded mandates as the ‘sly means’ to offload its own policy responsibilities without paying the costs (Bennett, 2014; Hart and Welham, 2016). On the other hand, subnational governments can perceive this opportunistic behaviour as a threat to their financial sustainability and their capacity to implement regional policies efficiently. In the United States, this discontent prompted President Clinton to pass the Unfunded Mandate Reform Act in 1995, which aimed to ‘end the imposition [. . .] of mandates [. . .] without adequate funding’ (Zelinsky, 1997). Rather than eradicating them, this law only slowed their approval and did not reverse unfunded mandates that were already in place (Bennett, 2014).
However, unfunded mandates are not unique to the United States, nor are they a feature of federal or developed countries only. For instance, McCarten (2003) illustrates how the Indian central government has gained control and underfunded Indian states’ budgets. In South Africa, many subnational authorities lack the financial and human capacities to implement essential local economic development policies (Khambule, 2020). Unfunded mandates are also a recurrent feature in decentralisation processes in developed countries. De Groot (2019), for example, explains how Dutch subnational administrations have struggled to cope with increasing responsibilities while resources remained stable. In Italy, the 2001 constitutional reform envisaged a better alignment of political authority and fiscal autonomy, but the law to implement this reform was only approved in 2009 and its implementation has been since postponed. This means that the mismatch between political and fiscal has not been corrected yet (Palermo and Wilson, 2014). And in the United Kingdom indirectly elected regional agencies remain expected to deliver on many fronts without the adequate budget (Lee, 2017; Morgan, 2002; UK Government, 2022). Political discourses on ‘levelling-up’ have re-awakened interest in this subject (Cörvers and Mayhew, 2021; UK Government, 2022). All this evidence points to the fact that unfunded mandates may exist in countries irrespective of their level of development. It raises the question of whether unfunded mandates impinge on development differently depending on the degree of decentralisation and on levels of development.
Why Do Unfunded Mandates Arise?
Unfunded mandates may stem from the balance in the degree of legitimacy of both central government and subnational entities in decentralisation processes. Rodríguez-Pose and Gill (2003) posit that whoever has the greatest legitimacy holds the upper hand in negotiations, decisively shaping the existence and dimension of unfunded mandates. When central governments hold the upper hand in terms of legitimacy, they will prefer to transfer powers and responsibilities, while keeping a tight grip on resources. This will lead to the proliferation of unfunded mandates. Conversely, when regional demands have a strong legitimacy, they may be more capable of securing not only greater responsibilities but also more resources to fulfil those responsibilities. Unfunded mandates are therefore construed as the reflection of a power hierarchy born from variations in legitimacy between the centre and the regions. Unfunded mandates are more likely to emerge when the central government dominates decentralisation. However, this power balance can vary over time, as decentralisation is a process, not a one-time event. Therefore, unfunded mandates may appear without deliberate imposition as regional needs become costlier and finance does not follow accordingly. They may also emerge as the negotiating capacity of regions is reduced, be it due to a region’s smaller size and wealth, or due to the presence of political differences with the national government.
The presence of unfunded mandates and their dimension may result in dysfunctional decentralised systems. After all, unfunded mandates undermine the capacity of subnational governments to gather information, tailor policies to local preferences, and implement them adequately (Klugman, 1994; Prud’homme, 1995). Lack of resources may also stifle policy innovation and hinder productive efficiency as regions cannot invest, innovate, specialise, and compete adequately (Donahue, 1997; Oates, 1999). The resulting ineffectiveness of subnational authorities may dent trust in public authorities and lower participation, transparency, and accountability, all of which are institutions that influence economic growth (Rodrik et al., 2004). This is especially worrying for regional economic development, since regional authorities are often in charge of their own development strategies, frequently starved of resources by unfunded mandates (Basdeo, 2012; Khambule, 2020).
Yet, despite theories warning that most – if not all – benefits of decentralisation may be scrapped in the presence of unfunded mandates, no empirical study to date has analysed the impact of unfunded mandates on economic growth. Instead, most comparative research has focused on the impact of the degree of fiscal and political decentralisation on growth, making the critical assumption that finance follows function. Most current evidence of unfunded mandates comes from case studies identifying instances where finance did not follow function in developing countries, such as in Indonesia or Tanzania (Bahl and Martínez-Vázquez, 2013; Boex and Martínez-Vázquez, 2006). Albeit trail-based, the results of these studies cannot be generalised. Hence, mapping the presence and dimension of unfunded mandates and empirically testing for their effect on economic performance remains crucial to understand the economic impact of decentralisation better.
Methodology
Research Question and Hypotheses
To determine to what extent unfunded mandates – defined as the mismatch between the powers and resources devolved to subnational elected tiers of government – affect economic growth in OECD regions and analyse the extent to which variations in the degree of decentralisation and in development moderate the potential growth impact of unfunded mandates, we formulate, based on the above theoretical discussion, the following hypotheses. These are displayed in a graphic manner in Figure 1:
Hypothesis 1: Holding everything else constant, the presence and dimension of unfunded mandates undermines economic growth.
As the effects of unfunded mandates may vary depending on levels of political decentralisation, fiscal decentralisation, and stages of development, we hypothesise:
Hypothesis 2a: As the dimension of unfunded mandates increases, regional economic growth is more negatively affected in regions with a higher level of political decentralisation, as the gap between responsibilities and resources will be higher there.
Hypothesis 2b: As the dimension of unfunded mandates increases, regional economic growth is less negatively affected in regions with a higher level of fiscal decentralisation, as the gap between responsibility and resources will be lower there.
Hypothesis 2c: As the dimension of unfunded mandates increases, regional economic growth is more negatively affected in poorer regions, as autonomous governments will have fewer resources at their disposal.
Data
Research on decentralisation has long suffered from a shortage of adequate country-level data. This problem is exacerbated when the analysis is conducted at subnational level (Martínez-Vázquez et al., 2017). Testing our hypotheses, however, requires regional data for all independent, dependent, and control variables over a reasonably large time period. Hence, one of the contributions of this research is putting together a large dataset containing information on the degree of decentralisation and the dimension of unfunded mandates for 518 regions in 30 OECD countries. The dataset covers, in an unbalanced way, the period between 1997 and 2018. The data are extracted, depending on availability, mainly from national and regional statistical offices (see Tables A1 and A2 in the Online Appendix) and complemented with data from international organisations. The extensive scope of this study in terms of regions and years, coupled with a lack of a single database compiling the data of interest, explains the slightly unbalanced nature of the panel.
The term ‘regions’ refers to the subnational tiers of government with sufficient data availability to measure the variables of interest (see Table A3 in the Online Appendix for the regions included in the analysis). The decision to study regions in the OECD responds to practical reasons of data availability. It also simplifies the task of controlling for unobserved heterogeneity and avoiding omitted variable bias, thanks to the relatively similar characteristics of most OECD countries (Ezcurra and Rodríguez-Pose, 2013).
Measuring fiscal and political decentralisation remains a highly contentious (Martínez-Vázquez and McNab, 2003). While the share of total public expenditure spent by subnational governments remains the most commonly used proxy for fiscal decentralisation in cross-country studies, it is only available at the national level. Therefore, we use the per capita expenditure capacity of each of the 518 regions as a proxy for their degree of fiscal decentralisation. This indicator is not available in international databases, meaning that the data collection involved checking the budgets for each of the 518 regions individually. Total regional public expenditure comprises all expenses undertaken by a particular subnational authority, irrespective of how they are funded (be them through own-source revenues, shared ones, or transfers).2 Values have later been divided by population and rendered comparable, converting them to constant 2015 USD, adjusted by purchasing power parity (see Table A1 in the Online Appendix).
Political decentralisation is too elusive a concept for it to be captured with a single measure. Scholars have created different indexes that aggregate various measures into an overall score that denotes the overall level of political decentralisation. Due to the regional-level data requirement of this study, we use the regional authority index (RAI) calculated by Hooghe et al. (2016, 2021), as it is the only index capturing within-country regional differences and the best at including a large variety of factors (Ezcurra and Rodríguez-Pose, 2013; Filippetti and Sacchi, 2016; Lessmann, 2012). The RAI overall score results from the aggregation of the values in eight sub-categories that are grouped under two main pillars: self-rule and shared rule.3 The former estimates the degree of authority exerted by the region over its territory, while the latter calculates a region’s influence over central government decisions. To avoid collinearity, we recalculate the index excluding the indicators related to fiscal decentralisation (fiscal autonomy and fiscal control) from the RAI overall score and use the resulting values as a proxy for political decentralisation at the regional level.
We use the above data to measure unfunded mandates. As the imbalance between power and resources, we make the variables for fiscal and political decentralisation comparable by standardising both with a mean value of 0 and a standard deviation of 1. We then subtract the values of fiscal decentralisation from the values of political decentralisation obtaining a relative index of unfunded mandates.4 With this conversion, we measure which regions have a larger or smaller unfunded mandate depending on whether their value is above or below the mean of 0, respectively. This index does not provide an absolute value of unfunded mandates for each region. Rather, it offers an estimated degree of unfunded mandates for each region relative to the gap between political and fiscal decentralisation in all other regions in the OECD. This has the advantage of comparing regions with one another and establishing which regions have wider or narrower unfunded mandates relative to the rest of the sample.
Finally, following previous literature on the link of decentralisation and economic growth, several control variables are incorporated into the model to avoid inconsistent parameters (Canavire-Bacarreza et al., 2017). We include regional population and region size, as larger regions in terms of population and/or land may have further resources to exploit to deliver a better economic performance (Arzaghi and Henderson, 2005; Congleton, 2006). Similarly, the level of development of a region may affect its growth potential. We therefore add regional gross domestic product (GDP) per capita in the model. We also control for human capital, a fundamental driver of growth. We measure human capital using the share of individuals in a region between the ages of 25 and 34 with a completed secondary education degree (Canavire-Bacarreza et al., 2020). Government size, often correlated with declines in economic growth (e.g. Afonso and Furceri, 2010), is measured as the total general government spending as a percentage of GDP and added in the full regressions (see Tables 2, 3, and A7 in the Online Appendix). Table A1 in the Online Appendix offers the description and sources of all variables above.
Model Specification
To test our hypotheses with our original panel dataset, a static panel-data region and time fixed-effects model is estimated with heteroscedasticity-robust standard errors (Newey and West, 1994). In this study, the baseline model adopts the following form:
(1)
where
represents annual GDP growth in region i for year t;
stands for unfunded mandates and denotes the difference between political decentralisation and fiscal decentralisation;
depicts the degree of fiscal decentralisation;
captures the level of political decentralisation;
encapsulates the relevant control variables (population size, level of development, educational attainment of young adults, region size, and national government size); µi and
are the region and time fixed-effects, respectively; while
denotes the error term.
Fixed-effects models have been normally used to estimate the economic impact of decentralisation in analyses of long-term decentralisation processes. However, one of the main problems of fixed-effects specifications when dealing with decentralisation is linked to the limited change over time of some decentralisation variables and, in particular, of political decentralisation. Fixed-effects models can also not consider time-invariant factors, such as region size or the presence of a particular region in a given country or continent. Random-effects estimators allows for both time- and time-invariant regressors, but have the drawback that region-individual effects can be correlated with some independent variables, leading to inconsistent coefficients (Hausman, 1978).
We, therefore, resort to Hausman–Taylor (HT) estimators (Hausman and Taylor, 1981) as our econometric approach. The use of HT, on the one hand, allows to calculate consistent coefficients for time-variant variables using their within-transformation as in the fixed-effects model; on the other hand, HT can also estimate coefficients for the time-invariant regressors. HT classifies variables as exogenous (i.e. correlated with the disturbance term only) or endogenous (i.e. correlated with the region-specific individual effects only), thereby partially controlling for endogeneity, a common concern in the literature (Baltagi et al., 2003; Canavire-Bacarreza et al., 2020). It also uses the between variation of time-variant exogenous regressors to derive internal instruments and hence does not require an additional external instrumental variable (Baltagi and Liu, 2012; Mitze, 2009). This is a key advantage because few strong instrumental variables exist for national-level studies, and this scarcity is aggravated at the regional scale.
The main model is specified as follows:
(2)
where,
captures annual GDP growth per region and acts as the dependent variable;
includes the time-variant exogenous control variables on regional population, level of development, education, and national government size. It also contains year dummy variables;
comprises three time-variant endogenous independent variables estimating the degree of unfunded mandates, the level of fiscal decentralisation, and that of political decentralisation, respectively;
represents the time-invariant exogenous variables and includes a series of supraregional dummy variables as well as regional size; µi denotes the fixed-effects term, while
stands for the disturbance term.
To test hypothesis 2, we also seek to determine the potentially mediating effect of a region’s level of development on the relationship between unfunded mandates and regional economic growth. Hence, following Lessmann (2012) and Filippetti and Sacchi (2016), we estimate an extended version of equation (2), where
includes three interaction terms between the variable for unfunded mandates and fiscal decentralisation, political decentralisation, and level of development, respectively.
Finally, we are not oblivious to discussions about reverse causality. It could be the case that lower economic growth spurs unfunded mandates, instead of the other way around. We therefore run a series of robustness checks. Due to the lack of appropriate external instrumental variables, model 2 is transformed from a static into a dynamic panel-data system-generalised method of moments (GMMs) (Arellano and Bover, 1995; Blundell and Bond, 1998; Roodman, 2009). This model is further discussed in the robustness checks section and its equation reads as follows:
(3)
where
captures the influence of past regional GDP growth on the next year’s growth rate, and the rest of variables are as stated above. The variables for unfunded mandates, and fiscal and political decentralisation are classified as endogenous in all system-GMM regressions.
Results
Mapping Unfunded Mandates
Decentralisation processes have become more common globally since the 1970s (Rodríguez-Pose and Gill, 2003). Table 1 shows the descriptive statistics for the three variables of interest for three different years within the 21-year period covered by the dataset.5 The means for fiscal and political decentralisation show a considerable overall increase of decentralisation across OECD countries since 1997. However, some regions in certain countries participate more in this trend than others, as the maximum value of fiscal decentralisation grows substantially over time whereas the minimum barely changes. For example, regional councils in New Zealand and French départements hardly increased their fiscal and political decentralisation; conversely, Estonian and Slovenian regions experienced fiscal increases of up to 313% and 198%, respectively. Figures A1 and A2 in the Online Appendix show the regional mean score over 21 years for both fiscal and political decentralisation, respectively.
| Variable | Year | Mean | Std. dev. | Min. | Max. |
|---|---|---|---|---|---|
| Fiscal decentralisation | 1997 | 3269.069 | 2558.776 | 43.01164 | 12277.95 |
| 2007 | 4495.101 | 4480.01 | 74.63864 | 37758.78 | |
| 2018 | 5613.333 | 5629.48 | 147.0289 | 51684.8 | |
| Political decentralisation | 1997 | 11.93552 | 6.729218 | 1 | 22 |
| 2007 | 12.45446 | 6.322351 | 1 | 22 | |
| 2018 | 12.73069 | 6.241244 | 2 | 23 | |
| Unfunded mandates | 1997 | 0.9309916 | 0.873728 | –1.318581 | 2.511998 |
| 2007 | –0.0015037 | 1.007628 | –2.885196 | 1.983407 | |
| 2018 | –0.0509553 | 0.911043 | –2.144672 | 1.861805 |
Author’s elaboration using data sources in Table A1.
Our main interest lies in measuring unfunded mandates. Table 1 above shows a reduction of its mean values over the 21-year period. This is an indication that fiscal decentralisation tends to catch up with political decentralisation over time. Nevertheless, Figure 2 reveals that unfunded mandates are pervasive across many parts of the OECD. Since the unfunded mandate variable is standardised with mean 0, Figure 2 uses a three-coloured palette to illustrate the areas where unfunded mandates are more or less pronounced, with green tones for lower values of unfunded mandates, ochre ones for values around the mean, and brown for higher values. These are obtained by calculating the mean of unfunded mandates for the entire 21-year period.6
Figure 2 shows that the presence of unfunded mandates follows no clear pattern in terms of which region displays a higher value of unfunded mandates. We nevertheless identify some general trends based on the type of decentralisation and levels of development of different regions. Unfunded mandates also vary across continents. First, higher values of unfunded mandates mostly affect relatively highly centralised countries. This includes most regions of New Zealand, the Netherlands, France, Poland, and Slovakia. However, exceptions exist, with unfunded mandates being rather prevalent in most regions of Germany, Spain, and Mexico. In any case, unfunded mandates appear less common in federal countries, such as Canada, the United States, Australia, and Austria. Second, unfunded mandates are more prevalent in regions with relatively lower levels of development. This is the case for regions in Colombia, Poland, and Slovakia. Once again, a remarkable exception is Germany, which stays at relatively high values of unfunded mandates. Finally, unfunded mandates are more widespread in Europe and Latin America than in Canada, the United States, and Australasian OECD countries.
In short, Figure 2 shows that unfunded mandates are common, but they follow no clear-cut pattern. Moreover, while there is some internal variation within countries, differences in unfunded mandates within countries are generally lower than across countries.
How do unfunded mandates and differences in the degree of political and fiscal decentralisation connect with regional economic growth? Figures 3–5 plot single-factor correlations between the three variables of interest and the dependent variable. Only fiscal decentralisation seems to be negatively correlated with economic growth, yet the correlation is very weak (Baskaran and Feld, 2013; Rodríguez-Pose and Ezcurra, 2011). The other two graphs reveal a practically inexistent relationship, which would support previous studies that have reported that fiscal and political decentralisation do not exert a statistically significant effect on economic growth (e.g. Thornton, 2007). Nevertheless, note that these are individual correlation graphs that do not control for other factors and thus cannot be used to infer a sound (non-)relationship between variables, especially for the multifaceted process of decentralisation. An inferential approach may change the initial conclusions drawn from these correlations.
Regression Results
Baseline Model: Two-Way Fixed-Effects Estimator
As stated in the ‘Methodology’ section, we fit a panel-data two-way fixed-effects estimator as the baseline model presented in equation (1).7 Table 2 displays the results using a forward-selection procedure that adds control variables sequentially. The number of observations is higher (over 6000 in 80% of the specifications) than in most previous analyses of decentralisation and the number of regions (502) only declines after introducing education and national government size.8 Regardless of the number of control variables, the independent variables of interest maintain the same coefficient signs in each of these estimations. Fiscal decentralisation has a negative connection with economic growth, whereas the reverse is true for political decentralisation, which boosts economic growth. This is in line with previous studies arguing that while political decentralisation may be beneficial for economic performance, fiscal decentralisation is connected with a loss of economic efficiency and growth (Rodríguez-Pose and Ezcurra, 2011). Most importantly, the presence of unfunded mandates is negatively associated with economic growth at over the 5% level of statistical significance. Hence, irrespective of the degree of fiscal and political decentralisation, the mismatch between the two undermines economic growth. Wider unfunded mandates cut regional economic performance. This implies that more decentralisation is not necessarily always desirable if the powers and resources at the disposal of regional governments are not properly aligned (Filippetti and Cerulli, 2018). This conforms with our main hypothesis based upon theories that underline the importance of equipping devolved authorities with the appropriate resources to reap the purported benefits of decentralisation (Prud’homme, 1995).
| Dependent variable | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|
| Regional GDP growth | FE | FE | FE | FE | FE |
| Unfunded mandates | –2.646** | –2.713** | –2.547** | –2.557** | –1.214** |
| (1.076) | (1.09) | (1.064) | (1.108) | (.502) | |
| Fiscal decentralisation | –1.853*** | –1.809*** | –3.502*** | –3.664*** | –2.01*** |
| (.69) | (.697) | (.751) | (.751) | (.357) | |
| Political decentralisation | 3.475** | 3.664** | 3.732*** | 3.352** | 1.508* |
| (1.441) | (1.46) | (1.423) | (1.465) | (.917) | |
| Population | –4.405*** | .063 | –3.879* | –2.299 | |
| (1.363) | (1.727) | (2.061) | (2.138) | ||
| Regional GDP pc | 12.877*** | 13.942*** | 13.663*** | ||
| (1.215) | (1.296) | (1.307) | |||
| Education | .082*** | .096*** | |||
| (.018) | (.018) | ||||
| National government size | –.123*** | ||||
| (.031) | |||||
| Year dummies | YES | YES | YES | YES | YES |
| Observations | 6788 | 6786 | 6786 | 6028 | 5923 |
| Number of regions | 502 | 502 | 502 | 476 | 475 |
| R2 | .188 | .19 | .235 | .264 | .268 |
| F-stat | 41.541 | 46.769 | 49.867 | 54.112 | 51.33 |
| F-stat (p-value) | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 |
Standard errors are in parentheses.
***
p < .01, **p < .05, *p < .1.
The magnitudes of the coefficient of the three variables of interest are only marginally sensitive to the inclusion of control variables. The different control variables display the expected values too. For both the level of development and the level of education, there is a highly statistically significant and positive association with growth. Conversely, the overall population and the national government size are negative and statistically significant, although in the former case to different degrees and with some marginal variation.
Main Model: HT Estimator
It could be argued that the two-way fixed-effects model is not ideal as it does not allow for the inclusion of time-invariant regressors. To consider both time-variant and time-invariant variables, some authors have resorted to random-effects estimators (e.g. Lessmann, 2012). Nevertheless, this approach is far from satisfactory, as the use of random-effects models may lead to inconsistent estimates in cases where region-specific individual effects are correlated with at least some independent regressors (Baltagi et al., 2003). We, therefore, use an HT model (Hausman and Taylor, 1981), which circumvents the ‘all or nothing’ dichotomy between fixed- and random-effects models, allowing for the inclusion of both time- and time-invariant regressors.
Table 3 presents the results of this estimation, following equation (2). The HT model classifies the variables for unfunded mandates, fiscal decentralisation, and political decentralisation as time-variant endogenous regressors. In all regressions, time-invariant supraregional dummies have been included for the different continental regions, namely, Western Europe, Eastern Europe, Asia Pacific, North America, and Latin America, the latter being the reference category. As in the previous table, control variables are added sequentially.
| Dependent variable | 1 | 2 | 3 | 4 | 5 | 6 |
|---|---|---|---|---|---|---|
| Regional GDP growth | HT | HT | HT | HT | HT | HT |
| Unfunded mandates | –.787 | –2.451** | –2.323** | –2.562** | –2.571** | –1.812** |
| (.774) | (1.045) | (1.015) | (1.068) | (1.064) | (.678) | |
| Fiscal decentralisation | –.539 | –1.672** | –2.57*** | –3.111*** | –3.028*** | –2.322*** |
| (.479) | (.655) | (.649) | (.689) | (.684) | (.62) | |
| Political decentralisation | 1.546 | 3.374** | 2.921** | 3.002** | 2.989** | 1.772* |
| (1.058) | (1.404) | (1.372) | (1.412) | (1.407) | (1.283) | |
| Population | –.185*** | –.22** | –.258* | –.393*** | –.26* | |
| (.062) | (.099) | (.139) | (.147) | (.128) | ||
| Regional GDP pc | 6.33*** | 9.753*** | 9.274*** | 7.61*** | ||
| (.64) | (.853) | (.819) | (.738) | |||
| Education | .041*** | .038*** | .048*** | |||
| (.015) | (.014) | (.013) | ||||
| Region size | .485*** | .472*** | ||||
| (.147) | (.128) | |||||
| National government size | –.117*** | |||||
| (.028) | ||||||
| Year dummies | YES | YES | YES | YES | YES | YES |
| Supraregional dummies | YES | YES | YES | YES | YES | YES |
| Observations | 6788 | 6786 | 6786 | 6028 | 6028 | 5934 |
| Number of regions | 502 | 502 | 502 | 476 | 476 | 476 |
| Chi2 | 2904.46 | 1737.025 | 1539.815 | 1506.715 | 1518.541 | 1615.315 |
| Chi2 (p-value) | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 |
Standard errors are in parentheses.
***
p < .01, **p < .05, *p < .1.
The HT model reports similar results to those in the baseline two-way fixed-effects model. The variables of interest preserve their sign, magnitudes, and, in most cases, the 5% level of statistical significance. Again, fiscal decentralisation has a negative correlation with economic growth while political decentralisation has a positive one. As in the baseline model, the coefficient for unfunded mandates is negative and statistically significant. This is in line with our main hypothesis since our regressions indicate that wider unfunded mandates entail lower rates of regional economic growth.9 Misalignments between fiscal and political decentralisation may thus hamper economic efficiency and growth (Khambule, 2021; Prud’homme, 1995). If the magnitudes of the coefficients for fiscal and political decentralisation are broadly similar but with opposite signs, it could be argued that the presence of smaller or larger unfunded mandates is key in determining whether decentralisation has an overall positive or negative effect, holding everything else constant. Hence, what seems to matter is not so much the extent to which a region is fiscally or politically decentralised, but rather the level of mismatch between the two. These empirical results give credence to those who warn against unfunded mandates, but not merely for political or constitutional reasons (Bennett, 2014; Posner, 1998), also for economic ones. Unfunded mandates, which prevail in decentralisation processes across the world, seriously undermine the economic impact of transferring powers and resources to lower tiers of government. When subnational governments are suddenly expected to provide public goods and services previously conducted by national governments, but with fewer resources, the result is worse economic performance. The impact on growth will then depend on how large the unfunded mandate is: the bigger the gap between political and fiscal decentralisation, the higher the impact on growth. Therefore, the results in Table 3 provide sufficient evidence to confirm our main hypothesis according to which as the unfunded mandate increases, regional economic growth decreases, ceteris paribus.
The sign and significance of the control variables included in the baseline model remain stable in the main HT regressions. In addition, new time invariant variables are added in the HT model. Following Arzaghi and Henderson (2005), region size is included in regression 5 and displays a positive and statistically significant relationship with growth. Finally, the supraregional dummies – omitted from the table for better visualisation – control for continental factors that may have affected economic performance. Latin American regions in OECD member countries grew faster during the period of analysis than regions in other supraregional groupings. This concurs with the economic convergence literature.10
Interaction Models
The previous results support our main hypothesis, that unfunded mandates represent a serious drag on the economic impact of decentralisation. The question now shifts to our second hypothesis about whether this relationship is affected by the degree of political and fiscal decentralisation in a region and by its level of development. In line with our second hypotheses, larger unfunded mandates in more politically decentralised regions may reduce growth as a higher transfer of powers may increase the gap between responsibilities and resources for regional governments (H2a). In contrast, more fiscally decentralised regions may be less negatively affected by wider unfunded mandates, as they would have a larger fiscal capacity of manoeuvre (H2b). Finally, larger unfunded mandates in poorer regions can reduce economic growth more strongly than for richer regions, which may have institutional characteristics that mitigate that negative effect (H2c). These hypotheses are tested in Table 4 through an extension of the main model, which includes three interaction terms between the variable for unfunded mandates and three other regressors.
| Dependent variable | 1 | 2 | 3 |
|---|---|---|---|
| Regional GDP growth | HT | HT | HT |
| Unfunded mandates | –2.573** | –3.718*** | –2.101* |
| (1.066) | (1.224) | (1.136) | |
| Political decentralisation | 3.045** | 2.863** | 3.083** |
| (1.413) | (1.41) | (1.405) | |
| Unfunded mandates × political decentralisation | –.579*** (.186) |
||
| Fiscal decentralisation | –3.899*** | –3.188*** | –3.277*** |
| (.735) | (.701) | (.674) | |
| Unfunded mandates × fiscal decentralisation | .165** (.08) |
||
| Regional GDP pc | 10.519*** | 9.132*** | 9.755*** |
| (.929) | (.815) | (.866) | |
| Unfunded mandates × regional GDP pc | –.085** | ||
| (.042) | |||
| Population | –.37** | –.378** | –.368** |
| (.175) | (.147) | (.155) | |
| Education | .043*** | .041*** | .042*** |
| (.015) | (.015) | (.015) | |
| Region size | .498*** | .494*** | .474*** |
| (.171) | (.14) | (.159) | |
| Year dummies | YES | YES | YES |
| Supraregional dummies | YES | YES | YES |
| Observations | 6028 | 6028 | 6028 |
| Number of regions | 476 | 476 | 476 |
| Chi2 | 1508.933 | 1544.26 | 1491.324 |
| Chi2 (p-value) | 0.000 | 0.000 | 0.000 |
Standard errors are in parentheses.
***
p < .01, **p < .05, *p < .1.
The predicted marginal values are plotted in Figures 6–8.11 The solid lines depict the true predicted relationship between variables as per the values on Table 4, while the dotted lines compare it with a counterfactual situation where the interaction term is not added. Since margin plots cannot be created automatically for HT estimations using statistical software, we calculate the linear relationship based on minimum, average, and maximum values of the unfunded mandate variable for regions with low, average, and high values of the interacted variable (fiscal decentralisation, political decentralisation, and regional GDP per capita).12 These figures serve as a stylised illustration of the directions and steepness of the lines rather than a portrayal of the exact predicted values of regional GDP growth.
The interaction term for political decentralisation and unfunded mandates in regression 1 in Table 4 is negative and strongly statistically significant. In Figure 6, the green solid line shows a steep downwards slope, implying that for highly politically decentralised regions, increasing unfunded mandates lowers growth. This concurs with the idea that political power at subnational levels engenders positive stimuli for economic growth but that these may not materialise if the needed finance does not accompany devolved functions (Bahl, 1999; Rodríguez-Pose and Bwire, 2004). The steep slope signals that any positive impact of growth linked to political decentralisation may be outweighed by the losses of unfunded mandates. Unfunded mandates, by contrast, have a far lower impact on the growth of regions with low levels of political decentralisation. Therefore, the interaction term markedly reduces growth for highly devolved authorities with high unfunded mandates scores, confirming hypothesis H2a.
Regression 2 in Table 4 shows that even if both unfunded mandates and fiscal decentralisation reduce economic growth, the interaction term between them is positive and statistically significant. This implies that in highly fiscally decentralised regions, unfunded mandates may somewhat cushion the negative connection between fiscal decentralisation and economic growth, as indicated by the green lines in Figure 7. This confirms our hypothesis H2b as highly decentralised regions have a greater margin of manoeuvre when faced with unfunded mandates, being able to somewhat mitigate – but not neutralise – the negative effects of fiscal decentralisation. This is in contrast with the situation depicted by the red lines. These are considerably steeper, thereby indicating that less fiscally decentralised regions are more sensitive to unfunded mandates than those that are more fiscally decentralised. This confirms hypothesis H2b.
Finally, regression 3 in Table 4 shows a positive and statistically significant impact of the level of development (regional GDP) on growth. The interaction term between the two is, however, negative and significant whereas its magnitude is marginally small. In Figure 8, richer regions (i.e. green lines) grow more than poorer regions (i.e. red lines), especially when unfunded mandates are minimal. As the dimension of unfunded mandates grows, economic growth in all regions decreases irrespective of the levels of development. However, it does so at different rates. At high values of unfunded mandates, the difference between predicted values of the two red lines is smaller (i.e. –2.43) than the difference between values of the two green lines (i.e. –3.76). This indicates a marginally stronger downwards effect of the interaction term on richer rather than poorer regions with a large value of unfunded mandates. This rejects hypothesis H2c.
In sum, the interaction terms indicate that, first, regions with higher fiscal decentralisation have more margin of manoeuvre than the less fiscally decentralised ones when confronted with unfunded mandates. More fiscally decentralised regions can thus soften the negative impact of unfunded mandates on growth. Second, political decentralisation can contribute to economic growth, but large unfunded mandates weaken this positive effect (Bahl, 1999). Hence, further political decentralisation in a context of high unfunded mandates is not the solution to a malfunctioning decentralised polity; instead, resources should be devolved to balance out unfunded mandates. Finally, contrary to both established theories and our secondary hypothesis H2c, wealthier regions with large unfunded mandates grow less than poorer ones with the same level of unfunded mandates, albeit only marginally. Low unfunded mandates make it easier for decentralisation to elicit economic growth both in rich and poor regions – but particularly in rich ones.
Controlling for Endogeneity
Table A13 in the Online Appendix13 includes additional tests using different techniques, fitting both static and dynamic models with the aim of checking the robustness of our main model.
To control for potential endogeneity, a dynamic model is estimated. The ideal approach to control for a potential case of reverse causality is to use an external instrumental variable strategy. Instruments have been used in the decentralisation literature, such as a country’s legal origin, the ethnic fractionalisation index, and even land area (e.g. Canavire-Bacarreza et al., 2020; Lessmann, 2012). Nonetheless, it is up for debate whether these instruments meet the exclusion restriction according to which the instrument is not correlated with the error term and thus has no persistent impact on the dependent variable (Martínez-Vázquez et al., 2017). For example, using land area as an instrument for the relationship between decentralisation and economic growth is questionable, as geography is indeed an essential determinant of economic performance and could be endogenous (e.g. Alesina and Spolaore, 2003; Crafts and Venables, 2003).
If finding a robust instrumental variable at the country level is complicated, doing so at the regional data becomes practically unfeasible. Therefore, to account for endogeneity, the static model 2 is transformed into the dynamic panel-data GMM in equation (3). We follow Arellano and Bover (1995) and Blundell and Bond’s (1998) system-GMM, as it operates better than the difference-GMM when finite samples have a large n and a moderately small t (Blundell et al., 2001; Roodman, 2009; Windmeijer, 2005). A system-GMM adds additional moment restrictions and allows lagged first differences to be used as instruments in the levels equation, thereby correcting for any potential bias likely to emerge in difference-GMM. To restrict the proliferation of instruments, the second and third lags have been chosen for the endogenous variables (Rodríguez-Pose and Ketterer, 2012). While it has been found that the efficiency gains of two-step system-GMM estimators are minimal compared to one-step procedures, both are presented (Hwang and Sun, 2018).
The dynamic models, which are asymptotically robust to heteroscedasticity, show that the variable for unfunded mandates remains negative and statistically significant, especially in models that are lag-restricted and have fewer instruments. Fiscal and political decentralisation are also consistently negative and positive, respectively. They are significant at over the 1% level in regression 12. As for the controls, both levels of education and development remain broadly significant. However, geographical variables, such as population and region size become insignificant. This may be a consequence of lagging the dependent variable, whose coefficients remain positive and significant. In all dynamic regressions, we conduct the Arellano–Bond test for serial correlation and find that there is first but no second-order serial correlation, as indicated by the large p-values. This fails to reject the null hypothesis meaning that the residuals themselves are not serially correlated (Henderson, 2003). In short, both static and dynamic tests confirm the robustness of the results.
Conclusion
Decentralisation is an ongoing global process. Yet, empirical studies on decentralisation remain inconclusive about its impact on economic performance. By assuming that ‘finance follows function’ – that is, that devolved responsibilities are accompanied by the necessary resources to fulfil them – these studies neglect that this is rarely the case. Mismatches between powers and resources, also known as unfunded mandates, are the norm rather than the exception (e.g. Bahl and Martínez-Vázquez, 2013). To our knowledge, this represents the first cross-regional empirical study assessing the extent to which the (mis)match between fiscal and political decentralisation may impact economic growth.
Using an original dataset for 518 regions in 30 OECD countries over a 21-year period (1997–2018), our analysis supports the view that it is not so much the degree of fiscal and political decentralisation that matters for regional economic growth, but rather the dimension of the mismatch between the two. This association is negative, statistically significant, and robust to different econometric estimations. Moreover, we find that fiscally decentralised regions enjoy a larger margin of manoeuvre than the less fiscally decentralised ones when confronted with unfunded mandates due to their higher fiscal capacity; that political decentralisation may be positive for economic growth, but that any potential benefits fizzle out in the presence of unfunded mandates; and that richer regions in the OECD are more negatively affected by unfunded mandates than poorer regions – albeit only marginally.
These results indicate that the debate about the economic impact of decentralisation has possibly been misguided. It should not be just about whether more or less political and/or fiscal decentralisation is needed, but also about whether the resources at the disposal of subnational governments are sufficient to address the responsibilities they have to face. Therefore, if decentralisation is to help promote economic growth, we should focus not necessarily on whether more transfers of powers and resources to regional governments are needed, but on achieving a better decentralisation; that is, a decentralisation process that appropriately matches responsibilities with resources. Otherwise, if finance does not follow function, governments cannot plausibly claim to be ‘decentralising authority’; instead, they are ‘decentralising problems’, as unfunded mandates render governance systems dysfunctional and endanger economic growth by incurring in losses of efficiency and most likely undermining trust in institutions.
As any other study on this topic, our research is also subject to limitations. First, lack of data constrains our capacity to delve deeper into the impact of decentralisation on developing economies. Future research may reproduce this analysis on a larger database encompassing developing countries to test whether the results hold. Second, different definitions and measures of unfunded mandates may be developed and tested using different econometric techniques and models. Discerning whether unfunded mandates occur in one policy area or another may be key, as not all policy domains may contribute equally to securing economic growth. Finally, in light of the importance of institutions on for growth (Acemoglu and Robinson, 2012; Rodríguez-Pose, 2013), future studies could include institutional variables. Although unavailable for some OECD regions, adding the European Quality of Government Index (Charron et al., 2021) to a subset of EU regions seems promising in order to test whether the effects of unfunded mandates remain unchanged.
Despite these limitations and the road ahead for future research, this study pushes the boundaries of our knowledge about how decentralisation affects economic development. We do so by delving on the concept of unfunded mandates, a factor that had attracted considerable theoretical attention but that until now had been empirically neglected. We have demonstrated that a balanced decentralisation system is crucial to reap the purported benefits of decentralisation on economic growth. This is far from merely being an academic concern. Overcoming the well-documented rise of territorial inequalities and the resulting wave of geographical discontent necessitates sound regional development strategies (Iammarino et al., 2019; Rodríguez-Pose, 2018). The effectiveness of these policies could be undermined by dysfunctional decentralisation processes When transferring authority to subnational tiers of government, decision-makers shall be better off building decentralisation systems that guarantee the fulfilment of the ‘finance follows function’ rule and avoid the proliferation of unfunded mandates, which can put a brake on the economic development prospects of many regions across the world.
Acknowledgments
The authors are grateful to participants at seminars in London and Madrid for comments and helpful feedback. The opinions versed in this article reflect those of the authors alone and do not necessarily represent the official opinion of the OECD.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research was conducted with the support of the LSE staff research fund.
ORCID iD
Andrés Rodríguez-Pose https://orcid.org/0000-0002-8041-0856
Footnotes
1. Other nomenclatures include ‘cost shifting’ (Sansom, 2009) and ‘service responsibility downloading’ (McMillan, 2006).
2. Similar to OECD standards, total expenditure data include current expenditure (employee compensation, social expenditure, subsidies, and other current transfers) as well as capital expenditure (OECD, 2021).
3. The eight sub-categories are institutional depth, policy scope, fiscal autonomy, representation, law making, executive control, fiscal control, and constitutional reform. The variety of indicators can explain why some maintain that the RAI accounts for both political and administrative decentralisation (Filippetti and Sacchi, 2016). Administrative decentralisation is not explicitly examined here due to lack of adequate regional-level data.
4. The same procedure has been followed using RAI’s fiscal autonomy and fiscal control values as our fiscal decentralisation values. Tables A8, A9, A10, and A14 in the Online Appendix show the regression results for all models calculated in the ‘Regression Results’ section. No significant variation of results is in evidence.
5. Tables A4 and A5 in the Online Appendix provide a complete set of descriptive statistics and pairwise correlations, respectively.
6. Calculating the mean may slightly neutralise variations that may occur from year to year. Figure A3 in the Online Appendix shows the values of unfunded mandates for 2018, which are virtually identical to those in Figure 2.
7. Due to the relatively short time span and the large number of regions, this should prevent non-stationarity from biasing the estimates by means of spurious correlations (Crescenzi and Rodríguez-Pose, 2012; Wooldridge, 2002). Augmented Dickey–Fuller and Phillips–Perron unit root tests are conducted in Table A6 in the Online Appendix and reject the null of non-stationarity at the conventional levels of significance.
8. Statistical estimations only consider regions with data for all variables inserted in the equation. Despite building a panel dataset with 518 regions that have data for at least one variable of interest, pooling them slightly reduces the number of regions covered. Regressions thus do not cover Icelandic and Estonian regions, the two regional councils of West Coast and Hawkes-Bay (New Zealand), and Panevėžys county (Lithuania). For regressions including the control variable for education, nine Colombian departments, and all Korean regions have been omitted due to lack of comparable data.
9. This is also true if the same estimations are run with regions that have average or higher levels of unfunded mandates (see Table A11 in the Online Appendix).
10. Table A11 in the Online Appendix also adds an Anglo-Saxon federal dummy variable resulting in no relevant variation compared to our main estimations.
12. The average value is taken by calculating the average of the minimum and maximum values of the variable at hand. It is included in each interaction graph for comparison purposes.
13. Table A14 in the Online Appendix shows the same robustness checks with the unfunded mandates variable calculated with scores from RAI, both for fiscal and political decentralisation.
References
Acemoglu D, Robinson JA (2012) Why Nations Fail. London: Crown Publishing Group.
Adler RW (1997) Unfunded Mandates and Fiscal Federalism: A Critique. Vanderbilt Law Review 50: 1137–1256.
Afonso A, Furceri D (2010) Government Size, Composition, Volatility and Economic Growth. European Journal of Political Economy 26 (4): 517–532.
Alesina A, Spolaore E (2003) The Size of Nations. Cambridge, MA: MIT Press.
Arellano M, Bover O (1995) Another Look at the Instrumental Variable Estimation of Error-Components Models. Journal of Econometrics 68 (1): 29–51.
Arzaghi M, Henderson JV (2005) Why Countries Are Fiscally Decentralizing. Journal of Public Economics 89: 1157–1189.
Bahl R (1999) Fiscal Decentralization as Development Policy. Public Budgeting & Finance 19 (2): 59–75.
Bahl R, Martínez-Vázquez J (2013) Sequencing Fiscal Decentralization. Annals of Economics and Finance 14 (2): 623–670.
Baltagi BH, Liu L (2012) The Hausman-Taylor Panel Data Model with Serial Correlation. Statistics & Probability Letters 82 (7): 1401–1406.
Baltagi BH, Bresson G, Pirotte A (2003) Fixed Effects, Random Effects or Hausman-Taylor? A Pretest Estimator. Economics Letters 79: 361–369.
Basdeo M (2012) The Impact and Dilemma of Unfunded Mandates Confronting Local Government in South Africa: A Comparative Analysis. Africa’s Public Service Delivery & Performance Review 1 (2): 51–66.
Baskaran T, Feld LP (2013) Fiscal Decentralization and Economic Growth in OECD Countries: Is There a Relationship? Public Finance Review 41 (4): 421–445.
Bennett JT (2014) Mandate Madness: How Congress Forces States and Localities to Do Its Bidding and Pay for the Privilege. New Brunswick, NJ: Transaction Publishers.
Blundell R, Bond S (1998) Initial Conditions and Moment Restrictions in Dynamic Panel Data Models. Journal of Econometrics 87 (1): 115–143.
Blundell R, Bond S, Windmeijer F (2001) Estimation in Dynamic Panel Data Models: Improving on the Performance of the Standard GMM Estimator. In: Baltagi BH, Fomby TB, Carter Hill R (eds) Nonstationary Panels, Panel Cointegration, and Dynamic Panels (Advances in Econometrics, Vol. 15). Bingley: Emerald Group Publishing Limited, pp.53–91.
Bodman P (2008) Fiscal Federalism and Economic Growth in the OECD. In: Future of Federalism Conference, Brisbane, QLD, Australia, 10–12 August.
Boex J, Martínez-Vázquez J (2006) Local Government Financial Reform in Developing Countries: The Case of Tanzania. New York: Palgrave Macmillan.
Canavire-Bacarreza G, Martínez-Vázquez J, Yedgenov B (2017) Reexamining the Determinants of Fiscal Decentralization: What Is the Role of Geography? Journal of Economic Geography 17 (6): 1209–1249.
Canavire-Bacarreza G, Martínez-Vázquez J, Yedgenov B (2020) Identifying and Disentangling the Impact of Fiscal Decentralization on Economic Growth. World Development 127: 104742–104756.
Charron N, Lapuente V, Bauhr M (2021) Sub-National Quality of Government in EU Member States: Presenting the 2021 European Quality of Government Index and Its Relationship with Covid-19 Indicators. The QoG Working Paper Series 2021:4. Gothenburg: University of Gothenburg. Available at: https://www.gu.se/sites/default/files/2021-05/2021_4_%20Charron_Lapuente_Bauhr.pdf (accessed 14 July 2021).
Congleton RD (2006) Asymmetric Federalism and the Political Economy of Decentralization. In: Ahmad E, Brosio G (eds) Handbook of Fiscal Federalism. Cheltenham: Edward Elgar Publishing pp. 131–153.
Cörvers F, Mayhew K (2021) Regional Inequalities: Causes and Cures. Oxford Review of Economic Policy 37 (1): 1–16.
Crafts N, Venables A (2003) Globalization in History: A Geographical Perspective. In: Bordo MD, Taylor AM, Williamson JG (eds) Globalization in Historical Perspective. Chicago, IL; London: University of Chicago Press, pp.323–369.
Crescenzi R, Rodríguez-Pose A (2012) Infrastructure and Regional Growth in the European Union. Papers in Regional Science 91 (3): 487–513.
De Groot H (2019) Growth and Shrinkage: Challenges for Governance and Solidarity. In: Bock B, de Groot H, Hospers G-J, et al. (eds) Towards a Cohesive Country: Population Decline and Regional Equality of Opportunity. The Hague: Platform, pp.30–49.
De Mello L, Barenstein M (2001) Fiscal Decentralisation and Governance: A Cross-Country Analysis. WP/01/71. Washington, DC: Fiscal Affairs Department, International Monetary Fund.
Donahue JD (1997) Disunited States. New York: HarperCollins.
Ebel R D, Yilmaz S (2002) On the Measurement and Impact of Fiscal Decentralization. Policy Research Working Paper Series 2809: 1–26
Ezcurra R, Rodríguez-Pose A (2013) Political Decentralization, Economic Growth and Regional Disparities in the OECD. Regional Studies 47 (3): 388–401.
Feld LP, Schnellenbach J (2011) Fiscal Federalism and Long-Run Macroeconomic Performance: A Survey of Recent Research. Environment and Planning C: Government and Policy 29: 224–243.
Filippetti A, Cerulli G (2018) Are Local Public Services Better Delivered in More Autonomous Regions? Evidence from European Regions Using a Dose-Response Approach. Papers in Regional Science 97 (3): 801–826.
Filippetti A, Sacchi A (2016) Decentralization and Economic Growth Reconsidered: The Role of Regional Authority. Environment and Planning C: Government and Policy 34 (8): 1793–1824.
Hausman J A (1978) Specification Tests in Econometrics. Econometrica 46 (6): 1251–1271.
Hausman JA, Taylor WA (1981) Panel Data and Unobservable Individual Effects. Econometrica 49 (6): 1377–1398.
Henderson JV (2003) The Urbanization Process and Economic Growth: The So-What Question. Journal of Economic Growth 8: 47–71.
Hooghe L, Marks G, Schakel AH, et al. (2016) Measuring Regional Authority: A Postfunctionalist Theory of Governance, Vol. I. Oxford: Oxford University Press.
Hooghe L, Marks G, Schakel AH, et al. (2021) Regional Authority Index (RAI) (EUI Research Data: Robert Schuman Centre for Advanced Studies). Available at: https://hdl.handle.net/1814/70298 (accessed 15 January 2021).
Hwang J, Sun Y (2018) Should We Go One Step Further? An Accurate Comparison of One-Step and Two-Step Procedures in a Generalized Method of Moments Framework. Journal of Econometrics 207 (2): 381–405.
Iammarino S, Rodríguez-Pose A, Storper M (2019) Regional Inequality in Europe: Evidence, Theory and Policy Implications. Journal of Economic Geography 19: 273–298.
Iimi A (2005) Decentralization and Economic Growth Revisited: An Empirical Note. Journal of Urban Economics 57: 449–461.
Khambule I (2020) A Question of Capacity and Funding: The Role of Local Economic Development Agencies in South Africa’s Development Landscape. Urban Forum 31: 95–113.
Khambule I (2021) Decentralisation or Deconcentration: The Case of Regional and Local Economic Development in South Africa. Local Economy 36 (1): 22–41.
Klugman J (1994) Decentralisation: A Survey of Literature from a Human Development Perspective. UNDP Human Development Report Office Occasional Papers 1994. Available at: http://dx.doi.org/10.2139/ssrn.2294658 (accessed 15 January 2021).
Lee N (2017) Powerhouse of Cards? Understanding the ‘Northern Powerhouse’. Regional Studies 51 (3): 478–489.
Lessmann C (2012) Regional Inequality and Decentralization: An Empirical Analysis. Environment and Planning A 44: 1363–1388.
McCarten JW (2003) The Challenge of Fiscal Discipline in the Indian States. In: Rodden J, Eskeland G, Litvack J (eds) Fiscal Decentralization and the Challenge of Hard Budget Constraints. Cambridge, MA: MIT Press, pp.249–286.
McMillan ML (2006) Municipal Relations with the Federal and Provincial Governments: A Fiscal Perspective. In: Young R, Leuprecht C (eds) Municipal-Federal-Provincial Relations in Canada. Montréal, QC, Canada; Kingston, ON, Canada; London; Ithaca, NY: McGill-Queen’s University Press, pp.45–82. Available at: https://www.queensu.ca/iigr/sites/webpublish.queensu.ca.iigrwww/files/files/pub/archive/SOTF/StateFed04.pdf (accessed 15 February 2021).
Martínez-Vázquez J, McNab RM (2003) Fiscal Decentralization and Economic Growth. World Development 31 (9): 1597–1616.
Martínez-Vázquez J, Lago-Peñas S, Sacchi A (2017) The Impact of Fiscal Decentralization: A Survey. Journal of Economic Surveys 31 (4): 1095–1129.
Mitze T (2009) Endogeneity in Panel Data Models with Time-Varying and Time-Fixed Regressors: To IV or Not IV? Ruhr Economic Paper No. 83. Available at: http://dx.doi.org/10.2139/ssrn.1358224 (accessed 15 February 2021).
Morgan K (2002) The English Question: Regional Perspectives on a Fractured Nation. Regional Studies 36: 797–810.
Musgrave RA (1959) The Theory of Public Finance: A Study in Public Economy. New York: McGraw-Hill.
Newey WK, West KD (1994) Automatic Lag Selection in Covariance Matrix Estimation. Review of Economic Studies 61 (4): 631–653.
Oates WE (1972) Fiscal Federalism. New York: Harcourt Brace Jovanovich.
Oates WE (1999) An Essay on Fiscal Federalism. Journal of Economic Literature 37 (3): 1120–1149.
OECD (2017) Multi-Level Governance Reforms: Overview of OECD Country Experiences. Paris: OECD Multi-Level Governance Studies, OECD Publishing.
OECD (2019) Making Decentralisation Work: A Handbook for Policy-Makers. Paris: OECD Multi-level Governance Studies, OECD Publishing.
OECD (2021) Regional Statistics. Available at: https://stats.oecd.org/Index.aspx?DataSetCode=REGION_DEMOGR
Palermo F, Wilson W (2014) The Multi-Level Dynamics of State Decentralization in Italy. Comparative European Politics 12: 510–530.
Posner PL (1998) The Politics of Unfunded Mandates: Whither Federalism? Washington, DC: Georgetown University Press.
Prud’homme R (1995) The Dangers of Decentralization. The World Bank Research Observer 10 (2): 201–220.
Putnam RD (1993) Making Democracy Work: Civic Traditions in Modern Italy. Princeton, NJ: Princeton University Press.
Rodríguez-Pose A (2013) Do institutions matter for regional development? Regional Studies 47 (7): 1034–1047.
Rodríguez-Pose A (2018) The Revenge of Places That Don’t Matter (and What to Do about It). Cambridge Journal of Regions, Economy and Society 11 (1): 189–209.
Rodríguez-Pose A, Bwire A (2004) The Economic (In)efficiency of Devolution. Environment and Planning A 36 (11): 1907–1928.
Rodríguez-Pose A, Ezcurra R (2011) Is Fiscal Decentralization Harmful for Economic Growth? Evidence from the OECD Countries. Journal of Economic Geography 11: 619–643.
Rodríguez-Pose A, Gill N (2003) The Global Trend Towards Devolution and Its Implications. Environment and Planning C: Government and 21: 333–351.
Rodríguez-Pose A, Gill N (2005) On the Economic ‘Dividend’ of Devolution. Regional Studies 39 (4): 405–420.
Rodríguez-Pose A, Ketterer TD (2012) Do Local Amenities Affect the Appeal of Regions in Europe for Migrants? Journal of Regional Science 52 (4): 535–561.
Rodríguez-Pose A, Sandall R (2008) From Identity to the Economy: Analysing the Evolution of the Decentralisation Discourse. Environment and Planning C: Government and Policy 26 (1): 54–72.
Rodrik D, Subramanian A, Trebbi F (2004) Institutions Rule: The Primacy of Institutions Over Geography and Integration in Economic Development. Journal of Economic Growth 9 (2): 131–165.
Roodman D (2009) How to Do Xtabond2: An Introduction to Difference and System GMM in Stata. The Stata Journal 9 (1): 86–136.
Ross JM (2018) Unfunded Mandates and Fiscal Structure: Empirical Evidence from a Synthetic Control Model. Public Administration Review 78 (1): 92–103.
Sansom G (2009) Commonwealth of Australia. In: Kincaid J, Steytler N (eds) Local Government and Metropolitan Regions in Federal Systems. Montréal, QC, Canada; Ithaca, NY: McGill-Queen’s University Press, pp.7–36.
Tiebout C M (1956) A Pure Theory of Local Expenditures. The Journal of Political Economy 64 (5): 416–424.
Thießen U (2003) Fiscal Decentralisation and Economic Growth in High-Income OECD Countries. Fiscal Studies 24 (3): 237–274.
Thornton J (2007) Fiscal Decentralization and Economic Growth Reconsidered. Journal of Urban Economics 61 (1): 64–70.
UK Government (2022) Levelling Up the United Kingdom. London: HM Government.
Windmeijer F (2005) A Finite Sample Correction for the Variance of Two-Step Linear Efficient GMM Estimators. Journal of Econometrics 126: 25–51.
Wooldridge JM (2002) Econometric Analysis of Cross Section and Panel Data. Cambridge, MA: MIT Press.
Zelinsky EA (1997) The Unsolved Problem of the Unfunded Mandate. Ohio Northern University Law Review 23 (3): 741–782.
Biographies
Andrés Rodríguez-Pose is the Princesa de Asturias Chair and a Professor of Economic Geography at the London School of Economics. He is the Director of the Cañada Blanch Centre LSE. He has a long track record of research in regional growth and inequality, fiscal and political decentralisation, institutions, discontent, and populism.
Miquel Vidal-Bover is a Junior Policy Analyst at the Centre for Entrepreneurship, SMEs, Regions and Cities, OECD. Previously, he was a Data and Research Specialist at the UN Women – Regional Office for the Arab States. He completed the MSc in Local Economic Development at the London School of Economics with the Best Academic Performance Award. His research interests cover decentralisation and territorial inequality.
Supplementary Material
Please find the following supplemental material visualised and available to download via Figshare in the display box below. Where there are more than one item, you can scroll through each tab to see each separate item.
Please note all supplemental material carries the same license as the article it is here associated with
