From Austerity to Expansion? Consolidation,Budget Surpluses,and the Decline of Fiscal Capacity

The Argument

At the core of the progressive consolidation view is a prediction about what will happen after a successful budget consolidation. This view sees austerity measures as a necessary but temporary adjustment to rebalance the budget and regain fiscal capacity. Afterward, it envisions a return to more ambitious fiscal policies. This argument assumes that postconsolidation policies are independent from preceding consolidation policies. The idea is that consolidation increases fiscal space without imposing restrictions on its use.

A similar view can be found in much of the work in comparative political economy. Studies taking budget balances or debt levels as explanatory variables for state activity rarely differentiate conceptually between deficits and surpluses but view them as symmetric. Starting with the often-confirmed result that fiscal stress triggers a contraction of public spending, which hurts different policy fields unequally, ⁵ these studies suggest that, by implication, abating fiscal stress will result in a renewed increase of those expenditures that were cut during consolidation. ⁶

We take issue with both approaches to postconsolidation fiscal policy. We argue that postconsolidation policies are neither independent from preceding consolidation policies, nor their mirror image. While rising fiscal stress indeed reduces fiscal capacity, declining fiscal stress does not necessarily lead to a reversal of budget trends. Instead, it is necessary to look at both the consolidation and postconsolidation periods and to investigate the mechanisms linking them.

Our argument takes up a set of core tenets of historical institutionalism, particularly its claims about path dependency, policy feedback, and policy legacies. ⁷ Following this literature, the argument is based on Schattschneider’s dictum that “new policies create a new politics.” ⁸ Today’s fiscal policies, we suggest, structure the conditions under which tomorrow’s fiscal policy decisions are made, even if their original causes have long disappeared. Explaining postconsolidation fiscal policy without taking into account what happened during consolidation is thus bound to fail because both processes are deeply intertwined. The measures taken to overcome fiscal crisis continue to structure the political, ideational, and institutional context in which fiscal policy is made afterward.

Examining the progressive consolidation view thus requires a long-term perspective. This limits the applicability of many studies on fiscal consolidation. These studies typically treat consolidation as a matter of technical adjustment and restrict their analysis of consolidation sustainability to a two-year or three-year window. ⁹ With such a short-term perspective, it is not possible to figure out how much discretion policymakers have after consolidation. Furthermore, this literature is almost entirely concerned with the causes or the macroeconomic effects of fiscal consolidations but not with their political effects. Hence, it has little to offer for an assessment of the progressive consolidation view.

Instead, we turn to the concept of a fiscal regime as developed by Paul Pierson. ¹⁰ He defines a fiscal regime as “the configuration of political interests, institutions, and policy arrangements that structure conflicts over taxes and spending. … At the height of a particular fiscal regime, the critical components—policies, politics, and institutions—will be mutually reinforcing.” ¹¹ At the core of a fiscal regime is a set of slow-moving macro trends all pointing in the same direction. Therefore, such a regime does not change frequently but persistently shapes the context in which fiscal policy decisions are being made.

A persistent transformation of the fiscal policy context is what lasting consolidation efforts are all about. Hence, successful austerity measures will typically involve a fiscal regime change. The core paradigms of fiscal policy must change to ensure that there is no return to deficit finance afterward. During such a transformation, all three dimensions of the fiscal regime—policies, politics, and institutions—are modified. These changes of the political setting persist beyond the consolidation itself.

Fiscal regimes are generally characterized by self-reinforcing effects; hence, a fiscal regime change occurs only in extraordinary circumstances. It will usually require the rare combination of a fundamental delegitimization of the old regime and the political will to force through comprehensive and unpopular reforms. This combination is often brought about by a severe fiscal crisis. The crisis may be the result of the gradual erosion or the rapid deterioration of public finances, for example, in the wake of a bursting bubble. In both cases, the old fiscal regime appears to be fundamentally exhausted and unable to respond effectively to the crisis. The collapse of the old regime is not only visible in deficit and debt ratios but is typically accompanied by high current account deficits, a falling exchange rate, spikes in interest rates, credit-rating downgrades, and other crisis symptoms. These dramatic events open a policy window for reforms that would have been inconceivable in pre-crisis times.

It has long been recognized that crises can trigger substantial policy reforms. ¹² However, a regime change goes deeper than mere policy change and requires a persistent change of paradigms. It sets in motion a process of ongoing policy reforms that does not stop when the crisis is over. To cause a fiscal regime change, a crisis has to be accompanied by a strong political commitment to consolidation. However, the commitment itself is a direct response to the fiscal crisis. As crisis and consolidation are intimately linked, disentangling their respective roles in regime change and identifying the degree of political choice is often very difficult. It is also beyond the scope of this paper, as we contend that far-reaching consolidation measures will affect all three dimensions of a fiscal regime and will thus always bring about the long-term effects we describe—independent from their original motivations. Therefore, the following analysis focuses on the consequences of regime change.

After a successful regime change, the new regime will be stabilized by self-reinforcing effects. First of all, these effects show up in specific policies. Under prolonged austerity, all public policy becomes “fiscalized”: whatever the substantial merit of a program, it is primarily evaluated by its budgetary impact. ¹³ In such a climate, all spending initiatives that might contribute to new deficits are judged critically. This particularly restricts investment, which used to be debt financed but has to be funded from current revenues in the new circumstances. Also, long-term commitments to new programs are hard to make, as they have an adverse effect on the budget outlook. These developments on the expenditure side are often supported by changes in the tax code, such as the abolishment of bracket creep or a shift from income taxation to indirect taxes on consumption. Such changes tend to make the tax code less progressive. They also put an effective lid on raising more revenue. Without bracket creep, an increase in revenue requires raising tax rates—something that is politically difficult at any time, and particularly after consolidation, when fiscal stress cannot be cited as an argument any longer.

The politics of fiscal policy are affected on three levels: First, the specific interests of important constituencies change when fiscal policies change. Second, consolidation reduces the political clout of some groups and strengthens others. Third, political parties adapt their electoral strategies to the new fiscal environment. The first effect arises because austerity affects the position individuals have in a political economy, for example, when the state pulls out of previously important policy fields. By reducing public sector involvement in the provision of certain goods—for example, by privatizing their production—the state weds their consumers to the market, to use Esping-Andersen’s memorable terminology. ¹⁴ When middle classes no longer receive essential services like health care or education through the public sector but purchase them on the market instead, disposable income becomes increasingly important to them. Hence, middle class constituencies become relatively more supportive of tax cuts compared to expenditure increases. The same mechanism holds for the privatization of public housing stocks, which makes new homeowners increasingly conscious of interest rates and affects their preferences regarding social spending. ¹⁵

The reason for the second effect is that austerity does not hit all budget areas equally. Some items are cut massively, while others suffer only symbolic retrenchment or none at all. This differential rate of contraction triggers a reconfiguration of the relative power resources of interest groups and policy coalitions. For example, public sector unions, a traditional stronghold of the labor movement, lose political influence when the public sector shrinks and fewer and fewer voters are employed directly by the state. More generally, tax-cut coalitions tend to benefit from consolidation, while spending coalitions are weakened as consolidation turns net beneficiaries of fiscal redistribution into net contributors. This affects both the defenders of existing spending programs and the proponents of new spending initiatives.

The third effect occurs because consolidation programs also have an impact on partisan politics and the strategic considerations of both government and opposition parties. Solving a debt crisis comes at great political cost as governments usually face vigorous resistance to cuts in public spending. At the same time, balancing the budget is a huge political victory, which is why successful consolidation governments are often reelected. ¹⁶ Consequently, reelected governments hesitate to risk their hard-won image as responsible guardians of the public purse once the crisis has been overcome. Furthermore, opposition parties are eager to present themselves as equally responsible as the government, in particular when they can be blamed for being in government during the preceding fiscal crisis. Thus, they are reluctant to demand a return to a more activist policy stance. This bipartisan commitment to tight fiscal policies is further underwritten by the general economic climate. Successful consolidations typically coincide with a general improvement of economic conditions: growth picks up, unemployment figures drop, and the current account deficit disappears. ¹⁷ While these economic achievements are as much a cause as an effect of the successful consolidation, they lend substantial political legitimacy to continuing the austerity course. ¹⁸

Besides policies and politics, a fiscal regime change also transforms fiscal institutions toward a more permanently austere design. Typical instruments are the introduction of a—politically or even legally binding—surplus target or a “debt brake.” Some countries also introduce legislated expenditure ceilings that limit the available financial resources for policymakers during the annual budgeting process. ¹⁹ These ceilings ensure that higher-than-expected revenues are not used for new spending. At the same time, they can be used for tax cuts and tax expenditures, as these do not count against the ceiling. Such reforms are often complemented by the creation of new authorities to enforce fiscal discipline, like fiscal councils. Finally, governments may resort to deliberately pessimistic forecasting methods to make sure that budget plans have only upside but no downside risks. ²⁰ These institutional changes remain in place when the consolidation is over. Although their purpose in times of consolidation is to force politicians’ hands to cut expenditures, they now bind politicians’ hands, preventing substantial new initiatives. Therefore, political space remains restricted, even if fiscal space increases.

Talking about a fiscal regime emphasizes that these transformations reinforce each other and become interlocked: institutional reforms affect the policies that are being introduced, new policies alter the composition of political interests, changing interest structures affect party strategies that in turn lead governments to adopt new institutional reforms, and so on. This does not mean that regime changes are irreversible. A regime can be brought down by exogenous shocks like a financial crisis or by endogenous erosion. For example, the economic rationale behind the regime may disappear when all public debt is paid down or when pre-saving for demographic change ceases to be necessary. Nevertheless, in the immediate aftermath of a successful consolidation, austerity will be firmly entrenched.

These theoretical considerations make us deeply skeptical about the predictions of the progressive consolidation view and its symmetric treatment of deficits and surpluses. Below, we substantiate this skepticism empirically. We analyze countries where consolidations were followed by a decade of almost uninterrupted budget surpluses. In principle, these countries had fiscal space for active policy measures, but they made little use of it and kept a tight lid on spending. Hence, the predictions of the progressive consolidation view do not hold. Just having a surplus does not change the underlying institutional and political commitments to austerity that made the surplus possible in the first place.

We argue that a specific variety of an austerity regime ²¹ arises from the ashes of the old regime, which we call a surplus regime. In the ideal typical surplus regime, attaining a budget surplus becomes the dominant goal of fiscal policy, and policymakers behave accordingly. The driving force behind this regime is a rejection of deficits and not an affirmation of surpluses. A regime change comes with an ideational resistance to deficit-financed spending within the polity. Macroeconomic priorities are adapted accordingly: Balancing the budget replaces balancing the economy as the most fundamental goal of fiscal policy. Annual surpluses become the new normal, and all political parties commit themselves to this goal. Parties may disagree about the use of the surplus, but they do not question its existence. The institutional reforms, political commitments, and ideational changes created during the consolidation remain in place, exactly because these reforms were instrumental in overcoming the deficit. Consequently, the fiscalization of public policy does not disappear, and new policy initiatives remain on the defensive.

Research Design and Case Selection

To examine the empirical validity of the progressive consolidation view, we analyze fiscal policy in developed economies between 1980 and 2009. We restrict our analysis to these three decades as they form the “age of permanent austerity,” in which fiscal policy faced permanent pressures from a maturing welfare state with an ageing constituency, international tax competition, and falling growth rates. This forced governments, “in fiscal terms…to run harder and harder just to stand still.” ²² In these circumstances, the progressive consolidation view seemed to promise much needed relief from growing fiscal pressure. The age of permanent austerity is thus an important scope condition to the argument developed above.

Within these three decades, we look at countries that ran substantial budget surpluses for an extended amount of time. ²³ These countries generated fiscal room for maneuver that was—at least theoretically—available for new policy initiatives. By analyzing these countries, we do not focus on the reasons for running surpluses but on the consequences of doing so. In other words, the endpoint of the consolidation literature, which tries to explain the success of budget consolidations, is our point of departure. We ask whether increasing fiscal space does indeed lead to a reassertion of fiscal capacity.

Surplus countries are ideal test cases for the progressive consolidation view. With regular surpluses and very low debt-to-GDP ratios, these countries were largely immune to financial market pressures often cited as a source of austerity policies. ²⁴ Furthermore, the available fiscal space allowed them to decide whether to cut taxes, increase spending, or repay the public debt and accumulate assets. The availability of fiscal space is the main reason why we do not analyze all consolidation episodes but only those leading to surpluses: a consolidation that just reduces the deficit to a lower number may indeed ease the pressure on fiscal space but will not increase it again. Even if the debt-to-GDP ratio and the relative interest burden shrink, additional room for maneuver is only created at a painfully slow pace. In contrast, a surplus generates “free cash flow” in the form of revenue not yet committed to specific expenditures and thus potentially available to reverse the decline of fiscal discretion. ²⁵ According to the progressive consolidation view, countries with extended surpluses should have increased discretionary spending quite substantially again.

However, not all surpluses indicate an increase of fiscal space. In many cases, they just result from an economic boom and growing tax revenue. When the boom ends, these surpluses disappear as quickly as they have come and pressures on fiscal space start to grow again. Hence, such surpluses are more akin to windfall profits. They do not arise from a fierce political struggle in which political and economic interests become invested in a new fiscal regime. Because they are not postconsolidation surpluses, they do not speak to the question of regaining fiscal capacity through political efforts.

Accordingly, we restrict our analysis to cases of extended surpluses, which we operationalize as countries running continuous surpluses for more than five years. This leaves us with the six cases presented in Table 1 (the Appendix shows the development of fiscal indicators year by year). ²⁶ Because these six cases developed almost in parallel during the second half of the 1990s, they are as causally homogenous as is possible in such comparisons.

OPEN IN VIEWER

Table 1.

Additional Fiscal Space in Six Surplus Countries in Percent of GDP.

			Net debt		Interest payments
Country	Surplus starts in	Average surplus	in t-1	in 2007	in t-1	in 2007
Australia	1998	0.86	21.2	−7.3	2.6	0.6
Canada	1997	0.98	68.5	22.4	5.3	0.6
Denmark	1999	2.39	35.1	−3.8	2.7	0.4
Finland	1998	3.48*	−14.6	−72.6	1.8	−0.6
New Zealand	1994	2.68	50.5	−5.0	3.2	0.1
Sweden	1998	1.23	24.6	−22.4	3.0	0.7
Average	/	1.94	30.9	−14.8	3.1	0.3

Source: OECD Economic Outlook.

*

Includes a social security surplus of 2-3 percent of GDP.

As Table 1 shows, fiscal space in these countries increased indeed substantially. Not only did they have the annual surplus available for political decision making, they also turned from a net-debtor to a net-creditor position. ²⁷ The resulting decline in the interest burden opened up considerable new space in the budget. How this space was used is the main question of interest here. Hence, we examine the development of public revenue and expenditure during the surplus period in the following analysis.

Proponents of the progressive consolidation view could argue that restricting the analysis to these six cases biases our results against the progressive view. After all, countries select themselves into our sample by running a restrictive fiscal policy. If countries were pursuing a more ambitious fiscal policy, they might not show up because they would either run deficits or quickly lose their surpluses again. In a sense, such a self-selection is exactly what we intend to show. One of our core claims is that the use of a surplus is not independent from its preservation. It is exactly this interconnection that we try to capture in the concept of a fiscal regime. Hence, such potentially excluded cases would serve to underline our point. If these countries were able to increase discretionary spending not by running balanced budgets, but by ceasing to do so, they would run contrary to the internal logic of the progressive consolidation view. According to this view, a linear and positive correlation exists between the budget balance and fiscal capacity. The progressive consolidation view holds that running a surplus creates a fiscal dividend, which can be spent in the following years—and that the options for spending the dividend are not restricted by the way the dividend was achieved. In contrast, those potentially excluded cases would show that it was necessary to forfeit consolidation successes in order to increase investment.

We begin our examination of the progressive consolidation view with a quantitative analysis of the six cases. In this analysis, we focus on their commonalities, not on their differences. We describe how major forms of expenditure developed during and after consolidation as well as what happened to revenue. The purpose of this section is to establish the distinctiveness of the spending patterns in the six surplus countries, but not to explain it. Such an explanation requires a more grounded historical analysis. Therefore, we elaborate on the quantitative results with an in-depth case study of the Swedish budget surplus. Sweden was governed by one of the main proponents of the progressive consolidation view, the Social Democrats, from 1994 until 2006. The Social Democrats designed the fiscal consolidation of the mid-1990s and had almost ten years to determine the allocation of the surpluses. Furthermore, Sweden is seen as the incarnation of the Social Democratic welfare state and has a strong tradition of an ambitious and interventionist fiscal policy with a high propensity to invest in public infrastructure, soft and hard. Sweden is thus a most-likely case ²⁸ for the progressive consolidation view.

Comparative Results

We start by describing the developments in the six surplus countries in very broad strokes. The main question is how different budget categories developed during consolidation and during the surplus. We first look at the consolidation periods preceding the surpluses.

All six countries experienced a fiscal crisis in the 1980s or 1990s: of the nineteen countries in the dataset, these were the only six countries that received a rating downgrade in this period ²⁹ and also the only countries that experienced a dramatic increase in the interest rates on government debt. ³⁰ In response, they embarked on fiscal consolidations that heavily focused on the expenditure side of the budget. Figure 1 shows that cyclically adjusted revenue and cyclically adjusted expenditure developed in markedly different ways: whereas revenue stayed almost flat during the consolidation, expenditure was cut massively. ³¹ The only exception to this pattern is Australia, where more than half of the consolidation was revenue driven. This was due to a change in government: the Labor government, in power until 1996, tried to balance the budget on the revenue side, while the incoming Liberal-National government focused on the expenditure side. All expenditure savings happened between 1996 and 1998.

OPEN IN VIEWER

Figure 1.

Cyclically Adjusted Revenue and Expenditure in Six Surplus Countries as Percent of GDP.

Total revenue and total expenditure are the most aggregate measures of budgetary developments. However, we suggested above that consolidations would affect different budget areas to a different extent. Following Castles, the most encompassing measure to test this proposition is net core expenditure: public expenditure net of interest payments minus social expenditure. ³² Figure 2 shows the respective shares of net core expenditure and of social expenditure in total net expenditure. On average, the share of net core expenditure declined during the consolidation period, confirming that the traditional welfare state was relatively protected on the aggregate level. ³³ Thus, the consolidations in surplus countries generally conform to the stylized facts of the empirical literature on successful budget consolidations: all except Australia were expenditure driven and all except Denmark cut core expenditure more than social expenditure.

OPEN IN VIEWER

Figure 2.

Share of Core and Social Expenditure in Net Expenditure in Six Surplus Countries.

However, our main concern is not the consolidation periods themselves but their aftermath. Therefore, we now examine the progressive consolidation view’s promise that fiscal capacity increases after a successful consolidation. As Figure 1 reveals, proponents of activist fiscal policy still faced an uphill battle: total expenditure kept falling during surplus years, even after excluding interest payments. ³⁴ Whereas the progressive consolidation view would predict that net expenditure converges upward toward gross expenditure with a falling interest burden, the opposite happened: gross expenditure converged downward toward a further falling net expenditure. Revenue also started to decline, indicating a political preference for tax cuts over discretionary spending. While it is not surprising that fiscal policy was kept tight during surplus years (otherwise there would have been no surplus), it actually tightened even further. This is confirmed by the development of core public spending. As Figure 2 shows, the share of core expenditure kept falling throughout large parts of the surplus period, with only a slight uptick at the end. After ten years of surpluses, core expenditure accounted for a lower share of net total expenditure than before consolidation.

These aggregate expenditure categories are still rather abstract measures of state capacity. It is therefore possible that they hide a shift in budgetary priorities toward investment. As Figure 3 shows, this view is somehow vindicated by a look at hard investment, measured as gross fixed capital formation, which rose slowly but steadily during the surplus years. Nevertheless, it took full ten years in surplus before investment had reached preconsolidation levels again. Moreover, this view is not borne out by soft public investment, which we measure as the sum of public spending on education, research and development, family policies, and active labor market policies. ³⁵ After a slight increase in the first surplus years, soft investment as a share of GDP started to decline again and after ten years in surplus it was lower than at the end of the consolidation period.

OPEN IN VIEWER

Figure 3.

Gross Fixed Capital Formation and Soft Public Investment in Six Surplus Countries as Percent of GDP.

So far, we have only looked at averages. However, these averages are not the random outcome of six divergent developments but usually a good representation of all cases. There is one important exception, which is hard public investment. The surge in gross fixed capital formation is predominantly driven by New Zealand, where its share of GDP almost doubled. One simple reason for this exceptional performance is a particularly high pent-up demand. New Zealand traditionally had the biggest public capital stock in the OECD and thus had correspondingly high replacement needs. Furthermore, the Labour government elected in 1999 put an explicit focus on raising public investment. This new investment, however, was not financed from the surplus but through an income tax increase legislated in 1999. ³⁶ Moreover, it still fell short of the economy’s needs: New Zealand’s government net capital stock is estimated to have declined from 102 percent of GDP in 1990 to 75 percent in 2001. ³⁷ Accordingly, New Zealand managers repeatedly identified “inadequate supply of infrastructure” as the greatest problem for doing business. ³⁸

The descriptive evidence presented so far contradicts the progressive consolidation view. However, proponents of this view could argue that the disappointing development may have nothing to do with the surplus. Instead, it could be caused by other factors like partisan preferences, globalization pressures, or slow economic growth. In other words, fiscal capacity might have suffered even more if there had been no surplus. It is thus necessary to control the development in surplus countries for these factors.

To control if spending patterns in surplus countries are indeed distinct, we perform a set of regressions with net core expenditure, gross fixed capital formation, and soft public investment as dependent variables. We use a panel of nineteen OECD countries between 1980 and 2009 and regress the annual change of the dependent variables on their lagged change and their lagged level, the budget balance of the previous year, a set of macroeconomic and political controls (see Table 2), and a lagged dummy variable for surplus years. We include this dummy variable to capture whether the link between budget balance and public spending in surplus countries is significantly different from the rest of the dataset. The specific model we estimate is the following:

Δ Depvar it = α t + β 1 × Depvar it − 1 + β 2 × Δ Depvar it − 1 + β 3 × Balance it − 1 + β 4 × Surplusdummy it − 1 + β 5 × Controls it + ε it

OPEN IN VIEWER

Table 2.

Definitions of Variables Used in Regression ³⁹ .

Variable	Definition
Netcore	General government public spending minus social expenditure minus interest payments as percent of GDP
Hard Investment	Gross fixed capital formation by general government as percent of GDP
Soft Investment	General government spending on education, research and development, family policies, and active labor market policies as percent of GDP
Cyc. Adj. Revenue	Cyclically adjusted current revenue of general government as percent of GDP
Balance	Budget balance of general government
Growth	Annual rate of GDP growth
Unemployment	Unemployment rate according to OECD definition
Left Government 40	Share of left parties in cabinet
Trade	Total trade (sum of imports and exports) as percent of GDP
Capital Openness	Index for the extent of openness in capital account transactions
Real Interest Rate	Long-term interest rates on government bonds
Population under 25	Share of population younger than 25

We include time-fixed effects in this model to control for the fact that fiscal policy in OECD economies has gradually become ever more austere between 1980 and the 2000s. As we argue that postconsolidation fiscal policy in surplus countries is very different from preconsolidation policy, we need to make sure that this is not just due to a common time-trend but caused by country-specific effects. Following Plümper et al., we do, however, not control for country-fixed effects. ⁴¹

The purpose of these regressions is neither to offer a full analysis of the determinants of public spending nor to evaluate the significance of specific explanatory factors. All we intend to show is that the failure of public investment to increase again cannot simply be explained by the standard competing independent variables. Even if these factors are included, fiscal capacity increases more slowly than expected.

The regression results are reported in Table 3. ⁴² The main variables of interest are the budget balance and the surplus dummy. First, and in line with the progressive consolidation view, there is indeed a significant positive effect of the lagged budget balance for all expenditure categories: the more positive (less negative) the budget balance, the more spending is increased as a share of GDP. Our main disagreement with the progressive view, however, is about the surplus dummy. According to the progressive view, deficit and surpluses have a symmetric impact on fiscal capacity; hence, the dummy-variable should not have any effect. Instead, we find significant negative coefficients for the surplus dummy with regard to both net core spending and soft investment. ⁴³ The effects are also quite substantial: for soft investment the effect of the surplus dummy is almost three times as big as the effect of an improvement of the budget balance by one percent of GDP. The effect for net core spending is even bigger. We also find a substantial effect for hard investment, which is, however, not significant. ⁴⁴ The main reason for this result seems to be the general decline of infrastructure investment in OECD countries during this period. On average, the nineteen OECD countries spent 2.98 percent of GDP for public investment in 1996, but only 2.72 percent in 2007. ⁴⁵ While surplus countries did not perform worse than this, they also did not perform substantially better.

OPEN IN VIEWER

Table 3.

Impact of the Budget Balance on Different Indicators of Fiscal Capacity.

	(1)	(2)	(3)	(4)
	Netcore	Hard	Soft	CA_Revenue
Balance(t-1)	0.171***	0.016***	0.043***	−0.052***
	(0.027)	(0.005)	(0.009)	(0.017)
Surplus dummy(t-1)	−0.653***	−0.038	−0.111*	−0.107
	(0.188)	(0.042)	(0.055)	(0.128)
Growth(t-1)	−0.025	0.019	−0.010	0.081*
	(0.061)	(0.010)	(0.016)	(0.034)
Delta Unemployment	0.218***	0.001	0.028	−0.192***
	(0.071)	(0.016)	(0.026)	(0.058)
Trade	−0.001	−0.001***	0.000	−0.000
	(0.001)	(0.000)	(0.001)	(0.001)
Capital Openness	−0.236*	−0.015	−0.009	−0.111
	(0.093)	(0.018)	(0.026)	(0.058)
Real Interest Rate	0.041	0.022*	0.005	0.060
	(0.057)	(0.011)	(0.017)	(0.032)
Left Government	−0.137	0.037	0.018	−0.020
	(0.144)	(0.030)	(0.057)	(0.108)
Left*Balance	0.043	0.013	0.001	−0.014
	(0.027)	(0.008)	(0.013)	(0.026)
Delta Netcore(t-1)	−0.294***
	(0.068)
Netcore(t-1)	−0.058***
	(0.021)
Delta Hard Investment(t-1)		0.026
		(0.073)
Investment(t-1)		−0.051***
		(0.017)
Delta Soft Investment(t-1)			0.019
			(0.061)
Soft Investment(t-1)			−0.034***
			(0.010)
Delta Population under 25			−0.131
			(0.100)
Delta Cyc. Adj. Revenue(t-1)				0.053
				(0.050)
Cyc. Adj. Revenue(t-1)				−0.008
				(0.007)
Constant	3.083***	0.340***	0.339	−0.342
	(0.441)	(0.085)	(0.180)	(0.319)
N	468	495	359	485
r2	0.351	0.229	0.392	0.316

Panel-corrected standard errors in parentheses. All regressions include time-fixed effects *p<0.05, **p<0.01, ***p<0.01.

We do not discuss the control variables in any detail, as their main purpose is to ensure that the negative correlation of surpluses and fiscal capacity is not just due to unobserved variation in these factors. One result that stands out with regard to these variables, however, is that we do not find a significant effect of government partisanship. Left governments neither seem to have a direct effect nor do they have an indirect effect through the interaction of partisanship and budget balance. This result is confirmed if we restrict the analysis to the forty-one surplus years in which left parties were in government. Even then we find a significantly negative effect for the surplus dummy. Thus, even left governments do not live up to the promises of the progressive consolidation view.

These results do not imply that budget balances have no effect on expenditure in surplus countries. After all, we also find a highly significant effect for the budget balance if we repeat the regression for surplus years only: the higher these countries’ surpluses, the more they spend in the following year. However, they still invest less than expected by the progressive consolidation view.

If surpluses were not used for higher spending, where did they go? To answer that question, we look at the development of cyclically adjusted revenue (Table 3, column 4). Here, we find again a highly significant effect of the budget balance: ceteris paribus, higher deficits are followed by higher revenue. In contrast to the expenditure categories, however, the surplus dummy is not significantly different from zero. Revenue behaves symmetrically during deficits and during surpluses. If anything, taxes are cut even more strongly than expected in surplus years. The tax-cutters’ equivalent to the progressive consolidation view is thus confirmed, again in line with the descriptive evidence presented above.

Overall, fiscal space, while created on the expenditure side, has mainly been used on the revenue side. Although governments did slightly reincrease public investment, there was no general reassertion of fiscal capacity. In other words, budgets reacted asymmetrically to the growth and decline of fiscal stress. States shrunk during consolidation and stayed lean during surplus. Taxes were cut as expected, but expenditure remained subdued to keep the budget in surplus.

Although the regressions confirm that fiscal policy in surplus countries is clearly different from other countries and that this is not just due to external factors, they do not explain this result. A dummy variable, after all, is largely devoid of explanatory content. The next step is thus to explain why fiscal capacity in surplus countries did not develop as predicted by proponents of the progressive consolidation view. This, however, is a task for which panel regressions are not suited very well. As our theoretical argument has stressed, fiscal regimes are characterized by historically contingent processes of path-dependency, self-reinforcement, and the deep interconnectedness of their individual elements. A method that focuses on annual changes thus faces inherent difficulties in dealing with such phenomena that only play out over the medium term. ⁴⁶ Instead, these processes call for a historically grounded approach. Therefore, we now proceed to examine one case of a surplus regime in greater detail.

The Swedish Case Up Close

As noted above, Sweden constitutes a most likely case for the progressive consolidation view. However, the quantitative results have already suggested that Sweden did not live up to this promise. The following case study deals with the political-economic processes and reforms behind this result. The analysis exemplifies how a surplus regime was created, how it has structured the country’s policies, politics, and fiscal institutions in the last two decades, and what its consequences are for the Swedish polity.

In Sweden, fiscal regime change occurred in the following sequence: First, a severe fiscal crisis put enormous pressure on political leaders to cut spending and implement far-reaching austerity measures. This crisis constituted a critical juncture in the development of Swedish fiscal policy. Second, Swedish politicians successfully pushed through a consolidation program. Thereby, new fiscal institutions were created and the political logic of a new fiscal regime was established. Third and finally, in the postcrisis years, the newly established surplus regime became embedded in the political discourse and the fiscal culture of the country.

Sweden experienced a severe multilayered crisis at the beginning of the 1990s. Public finances, the economy, the banking system, the exchange-rate regime, and the housing market were all in a state of emergency. Annual deficits reached double digits, and the public debt ratio almost doubled from about 45 percent of GDP in 1990 to over 80 percent in 1994. The whole political-economic system was at stake, and the entire population had an intense feeling of national crisis.

Political leaders in both political camps reacted with an extensive reform program to restructure public finances and get the financial and economic crisis under control. The consolidation measures comprised drastic expenditure cuts and targeted tax increases. By April 1995, the austerity measures added up to 8 percent of GDP. ⁴⁷ The crisis was the trigger that made a paradigmatic shift of the Swedish fiscal regime and a wider political turn in many policy areas possible. These changes affected all three dimensions of a fiscal regime.

In the dimension of policies, two major reforms during the 1990s have been particularly decisive in establishing the new fiscal regime: on the revenue side, the tax reform of 1990-91 and, on the expenditure side, the pension reform of 1994-98. The tax reform is often called the “tax reform of the century” because it was the most far-reaching tax reform in any industrialized country in the postwar period. ⁴⁸ Important features of the reform were the end of bracket creep, massive cuts of marginal income taxes, the introduction of dual income taxation of wages and capital, and a broadening of the tax base by eliminating loopholes and tax expenditures and by increasing consumption taxes.

Overall, the reform had regressive effects and lowered the tax yield. According to the official evaluation of the reform, it was underfinanced and at first exacerbated the fiscal crisis. ⁴⁹ Furthermore, it had far-reaching, long-term implications that have been manifesting themselves only gradually over time. In particular, the elimination of bracket creep may at first have seemed unimportant, but has had great impact on the development of public revenue in the long run.

Since the reform, Swedish tax policy has continued in the same vein. The Social Democratic government abolished the inheritance tax in 2004—after having reduced it before. In 1989, the last year before the great reform, this tax contributed 5 percent to the total receipts of the Swedish state. ⁵⁰ The new conservative government that took office in 2006 lowered tax rates further and abandoned some taxes completely like the wealth tax and the property tax. However, the tax relief never went so far as to threaten the budget surplus. For the election year 2014, the government has recently announced it will cut income taxes once again. It would be the fifth time since coming to power, and the tax relief would add up to 15 billion Swedish kronor.

On the expenditure side, public programs have increasingly become “fiscalized” and nearly all of the extensive cuts made during the crisis years have not been reversed ever since—although the crisis is long over and the fiscal space for new spending has been available for many years. ⁵¹ Government activities in general and new spending initiatives in particular have mainly been evaluated by their budgetary impact and not by other criteria as the great pension reform of 1994-98 exemplarily shows.

The pension reform established an entirely new pension system. A funded scheme was added to the widely reformed pay-as-you-go system. Since the reform, the pension system has been completely independent financially from the budget. The public purse no longer cross-subsidizes the pensions. Only the revenues raised by the pension system are paid out. If contributions decline, pension benefits will be automatically adjusted downward. Considering the demographic development of an ageing population, this is a very likely scenario. The main impetus behind the reform was to curb public spending and to establish a “mass investment culture” in Swedish pension savings. ⁵² The demands of pensioners were subordinated to a balanced budget. As a consequence, since the reform a vast majority of Swedes (roughly 90 percent) feel the need to complement their public pension with private insurance. ⁵³

The ongoing budget surpluses have changed Swedish politics and the debates on fiscal policy as well. It has become very difficult to support even a temporary return to deficit finance. Calls for increased public spending in line with the progressive consolidation view can hardly be heard in postcrisis Sweden. Running a surplus has acquired huge symbolic importance. All parties try hard to avoid the impression of behaving financially irresponsibly. Government and opposition are outbidding each other in a race to see who can spend less public money. This trend has intensified since debt crises have occurred in many other European countries. For instance, during the debates on the budget proposal for 2012, the Social Democratic shadow minister of finance, Magdalena Andersson, demanded even more fiscal restraint from the government. The Social Democrats made a budget proposal that would have saved an additional 5 billion Swedish kronor compared to the government’s budget. ⁵⁴

A new logic of blame avoidance appears to be at work in Swedish politics, where no politician wants to be judged as profligate. For the time being, the assumption that having a surplus is the hallmark of fiscal responsibility is not seriously contested in either political camp, the media, or the electorate. Running budget surpluses every year has become the new normal in Swedish fiscal policy.

Furthermore, public spending coalitions have been weakened over the last two decades. As part of the consolidation efforts, the number of people employed in the public sector declined from 1.3 million in 1990 to less than 1.1 million in 1998. ⁵⁵ The importance of labor unions has diminished ⁵⁶ and service provision has been privatized as the example of unemployment protection shows. Although times of fiscal crisis are long gone, the current conservative government cut measures for active labor market policies and unemployment benefits further. For the first time, spending on active labor market policy dropped below 1 percent of GDP in 2008 while the unemployment rate remains high. ⁵⁷ At the same time the government increased the employees’ contributions to the unemployment insurance (individual contributions tripled in 2007) and significantly lowered state subsidies and tax expenditures (before, 40 percent of the contributions could be deducted). ⁵⁸

As a consequence, the coverage of the unemployment scheme has been reduced, benefits vary more by income status, and the rate of people being members of unions’ unemployment funds dropped by 12 percent after the raise of contributions. ⁵⁹ This means that a large number of wage earners with a high risk of unemployment now lack sufficient protection in case of unemployment. ⁶⁰ Because the Swedish unemployment scheme is administered by the unions (Ghent system), these developments have made union membership less attractive. Low-income earners are often left without coverage because the contributions have become too high. High-income earners turn to supplementary private insurance because the benefits are too low. This is mainly caused by the de-indexation of the benefit’s ceiling. Although the official net replacement rate is 80 percent, the average worker de facto only receives around 50 percent due to the lowered ceiling. ⁶¹ The heavy reduction of governmental payments to the unemployment funds, the privatization of unemployment risks, and the reduction of public employment all undermine the Ghent system and the power of unions. This trend has been complemented by an encompassing reform of the administration of unemployment and the privatization of job centers in 2007-08.

These developments in Swedish public policies and politics are underpinned by a tight fiscal framework and fiscal institutions that foster budget discipline and strengthen the politics of running a surplus. Three major reforms stand out: the new budgetary process with an expenditure ceiling as the core element, the surplus target, and the prohibition of municipal deficits.

After the crisis, the Swedish budgetary process was fundamentally transformed. Until then, the budget had only covered the next fiscal year. However, since 1997, it has been planned for three years in advance and has become much more hierarchical (top-down budgeting). ⁶² The core element to enforce fiscal discipline is a rolling expenditure ceiling that sets binding limits for the nominal total government expenditure of the central state and the pension system. It is rolled each year, which means that—unless exceptional circumstances prevail—the ceilings for the first and the second year of the decision period are inherited from the previous year’s decision. ⁶³ In other words, the aggregate spending decisions have already been made when the annual budget is discussed.

Once the overall spending level is fixed, budgetary items can only be increased if others are lowered to the same extent. The parliament cannot adopt a budget bill with higher expenditures than the original proposal by the government. ⁶⁴ The expenditure ceiling includes all spending areas of the budget—except interest payments on government debt. They are excluded because interest rates are highly volatile and difficult to predict—especially in times of fiscal crisis. The effect is that a reduction of interest payments due to consolidation does not lead to more room for increased spending. However, taxes can be cut further, as the ceiling only affects expenditure but not revenue. Of course, the government can spend even less than stipulated by the ceiling. For example, the incoming conservative government lowered the existing ceiling after taking office in 2006. ⁶⁵

Another important institution in Swedish fiscal policy is the surplus target, which commits the Swedish central state to running an average budget surplus of 1 percent of GDP over the business cycle. The surplus target was introduced on a trial basis in 1997 and has been an official policy guideline for the government since 2000. In 2011, the surplus target was tightened once more and became legally binding. ⁶⁶ In practice, Swedish governments have even outperformed the surplus rules and saved more money than legally required. In the last decade, Sweden has not only generated an average surplus over the business cycle, but has had structural budget surpluses in every phase of it—even in times of recession.

The last fiscal rule that strengthens the surplus regime is the prohibition of municipal deficits for consumption. It was adopted in 2000 and requires all municipalities and counties (län) to restrict their expenses or to raise their revenues so they always show a balanced budget. ⁶⁷ These reforms have been embedded in a wider institutional configuration that supports a tight fiscal policy as well. Sweden joined the European Union (EU) in 1995 and has implemented all EU fiscal rules—although not joining the euro area. In 2007, a fiscal policy council was founded that advises the government and evaluates the state of public finances. Many fiscal policy decisions are now in the hands of experts and bureaucrats and can no longer be altered by elected politicians. A massive rebuilding of fiscal institutions has taken place in the last two decades. With such tight budgetary institutions, the Swedish government hardly has any discretion left for new spending initiatives. It is legally obliged to run a budget surplus in nearly all circumstances.

The analysis of the three dimensions of the Swedish fiscal regime has shown that the fiscal crisis at the beginning of the 1990s played a decisive role in triggering major reforms and austerity measures. However, it has also shown that austerity persisted long after the crisis was over and that policy, politics, and fiscal institutions continued in the same vein ever since—although fiscal pressure had disappeared. Many anticrisis measures were not reversed, and some were even tightened during the surplus period. We argue that continuing austerity can best be explained with the underlying logic and dynamic of a new durable fiscal paradigm that we call surplus regime.

This new regime has been established in Sweden since the great crisis of the early 1990s, with all its consequences for policies and society. Sweden has moved from a very expansionary fiscal regime to one of the most austere regimes in the developed world. All important dimensions of a fiscal regime were transformed: policies, politics, and institutions. These trends in the political setting are likely to remain powerful in the foreseeable future: fiscal regimes are difficult to build because they require investment of political capital and overcoming stakeholder resistance. They are also difficult to reverse because of self-reinforcing effects, the sluggishness of institutions, and legally binding rules. The fiscal environment in the Swedish surplus regime imposes hard constraints on government activity and strongly defines Swedish politics.

So far the toughest test of the persistence of the surplus regime has been the deep recession of 2009. As a small and open economy, Sweden was hit hard by the global financial crisis. The Swedish economy shrank by 5 percent in 2009, a sharper downturn than in any year of the 1990s crisis. ⁶⁸ However, even in these circumstances, sound public finances and not the stabilization of the economy remained the top policy priority of the government.

In 2009, Sweden experienced—for the first time in many years—a small budget deficit of less than 1 percent of GDP. The reasons for the deficit were lower revenue, due to the economic slump, and higher expenditure, due to automatic stabilizers. It was not primarily caused by a government-induced fiscal stimulus. The cyclically adjusted budget for 2009 showed a huge structural surplus. ⁶⁹ The government did not significantly alter its fiscal policy goals in response to the recession. It stuck to the surplus target and met the expenditure ceiling in every single year. Even the Swedish fiscal policy council criticized the government’s policy and stated:

It is our opinion that the expenditure ceiling should not be defended at any price during a deep recession. The expenditure ceiling has no value in itself. … If in a deep recession the regulatory framework instead limits the policy so that it is obviously ineffective, the short-term cost of keeping the ceiling, no matter what the economic situation, is too high. ⁷⁰

However, the path taken will most likely be continued in the future. The medium-term financial forecast of the Swedish government for the years 2013-2015 projected surpluses of up to 3 percent of GDP. The improvement of the budget balance would be achieved solely by expenditure cuts and not by revenue increases. ⁷¹ Since then, an economic downturn has rendered these projections somewhat moot. Still, it did not trigger any fundamental rethinking of fiscal priorities.

Two decades of surplus policy have led to visible political-economic consequences. Although the Swedish state levies more revenue than it spends each year, the public revenue ratio and the public expenditure ratio have been shrinking constantly since the crisis in the 1990s. Swedish government activity has steadily declined. The change of the fiscal regime is not only indicated by different policies, politics, and fiscal institutions, but by the spending numbers themselves, as Figure 4 shows. Public revenue and public expenditure have fallen by more than ten percentage points in the last twenty years. Expenditure has fallen even faster than revenue, thereby generating the surplus. This has particularly affected core public expenditure, which has fallen from more than 35 percent of GDP to less than 25 percent. Important indicators of fiscal capacity declined during consolidation and were not increased again in times of surplus. In particular, soft public investment has decreased more than in other advanced economies. ⁷²

OPEN IN VIEWER

Figure 4.

Public Revenue and Public Expenditure in Sweden as Percent of GDP.

Although the deepest cuts concerned core expenditure, the effects of the ongoing surplus regime did not spare the welfare state. The new fiscal regime has made it possible to retrench the universal and generous welfare state that was once described as a practically “immovable object.” ⁷³ The Swedish welfare state is still one of the biggest in the world and remains extremely popular, ⁷⁴ but it has been slowly and gradually transformed through a process of layering. ⁷⁵ Most social services like health care, childcare, elderly care, and education are still publicly funded, but private for-profit companies increasingly provide them. The government supports this development through tax expenditures and subsidies. This trend undermines universalism, and it tends to intensify social stratification because mainly the better offs rely on private providers. If it continues, the Swedish welfare system will move more and more in the direction of a liberal model with basic public security supplemented by private topping-up. ⁷⁶

The partial privatization of welfare services in combination with lowered taxes means a “great risk shift” ⁷⁷ from the public to the individual and has already affected the socioeconomic conditions in Swedish society. One of the most obvious developments is the dramatic rise in inequality. The Gini coefficient increased from 0.19 in the 1980s to 0.26 in 2008. In the last ten years, income inequality has risen faster in Sweden than in almost all other industrialized countries. ⁷⁸

Conclusion

The fiscal crisis of the United States and the euro area foreshadows a period of prolonged budgetary austerity. Whether austerity programs will be successful remains to be seen; and if they are, it is uncertain what the long-term impact on the role of the state in the economy will be. Reformers adopting the progressive consolidation view promise that painful reforms will help reinstall the state’s capacity for promoting economic welfare, but the empirical validity of this promise is unclear. The evidence presented here suggests that, even if (or especially if) anticrisis measures are successful on their own terms, there is reason to expect that their collateral damage to the expenditure side of the budget will remain in place.

This is not to deny the benefits of running surpluses: countries with surpluses pay very low interest rates and are relatively well prepared for the ageing of their populations. Some of them enacted big stimulus programs in the extraordinary circumstances of the financial crisis in late 2008—although not without hesitation. With regard to the progressive consolidation view, they have indeed been able to stop the drain on their investment expenditures. Although they were able to avoid further market-enforced retrenchment, they did not, however, systematically reincrease the activities of the state. The idea that surpluses are basically a mirror image of deficits proved to be unfounded. Instead, where countries decided to allocate surpluses to political initiatives, they primarily opted for tax cuts. Sweden is a prime example of these developments. The public spending ratio has been falling for two decades, the tax burden has been lowered, and even the universal and generous welfare state has not remained unaffected. Especially in the social service sectors, a clear trend toward privatization and liberalization can be observed.

Furthermore, the analyzed countries achieved their surpluses exactly by cutting those programs that are generally perceived as crucial from a social investment perspective. The consolidation efforts in the 1990s were expenditure-led in all six countries, with sometimes harsh cuts to core spending areas. Although the consolidation literature argues that expenditure-led consolidations are more likely to succeed than revenue-led consolidations, ⁷⁹ this article shows that this success comes at a price, and that euphoria about a growing capacity of the state is certainly unjustified. Our analysis reveals that budget surpluses and shrinking debt are embedded in the political and institutional context of a surplus regime, which makes an activist use of the newly opened fiscal space rather unlikely.

We have argued that this regime is based on self-reinforcing processes that strengthen its foundations over time. However, the longer such a regime exists, the more it will also produce self-undermining tendencies: the progressive commitment to surpluses may be questioned when the collateral damage of low public investment becomes increasingly visible. The conservative commitment may be challenged when all public debt is paid off and the state becomes an important player in the private economy by investing its surpluses in financial assets. This opens up two potential avenues for a breakdown of the surplus regime. Structural deficits may return when expenditure is finally raised or when tax cuts become too big. At this point, it is purely speculative which result is more likely. Yet, we would argue that the recent debt crises in several advanced economies have helped prolong the life span of surplus regimes at least over the medium term. As long as they remain entrenched, surpluses may not be a sign of the growing capacity of the state but rather of its shrinking ambition.