Economic Countercyclical Fiscal Policy and Growth in Democracies After the 2008 Recession

The 2008 financial crisis plunged the United States and the world into an economic downturn the likes of which had not been seen since the Great Depression. It was characterized by massive demand shocks, bank bailouts, widespread unemployment, and a very slow recovery for many economies worldwide. The debate has centered around how to expedite the recovery.

Internationally, and within the American political divide, the debate primarily exists between proponents of market-based procyclical economic policy and proponents of Keynesian countercyclical policy. Procyclical theories hold that government action cannot end crises in effective ways. The only thing to do in a downturn is to reduce regulations, taxes, and government debt to ensure that, when the recovery begins, it is free to make progress without the government hampering growth. Countercyclical theories assume that a government's best response to falling consumer spending is to step in and increase spending, thereby stimulating the economy. Now that the US has reached the recovery stage of this financial crisis, we can examine the evidence since 2008 to empirically test the validity of procyclical and countercyclical theories. This study will consider the efficacy of stimulus packages, a principal tool of recovery for countercyclical fiscal policy. In particular, whether countries that instituted major countercyclical policy initiatives, like stimulus packages, in response to the 2008 crisis fared better economically than those that did not. This study will first propose that the economics behind the countercyclical fiscal policy initiatives in democratic countries is sound, and second, that countries that instituted such initiatives will have higher GDP growth post-2008 than those that did not. To examine this claim, I will run a statistical regression on GDP growth using country-years in all democracies, as defined by Freedom House, assessing data from 2009 to 2012. I distinguish between those nations which have instituted or continued major countercyclical policy initiatives that year for the independent variable, and I include bank leverage, population growth, and GDP per capita as control variables. This study will first proceed to review existing literature on economic recoveries and the work of other scholars on the factors which may determine their success. Second, I will detail the underlying theory behind the factors I chose to examine and the causal narrative for each factor. Next, I will elucidate my hypothesis and detail the research design and methodology. Finally, the study will conclude with an analysis of the data and summary of findings.

Literature Review

Current gaps in the literature on economic recovery from financial crises arise from the present lack of major studies on the effectiveness of countercyclical fiscal policy. The lack of such work has been particularly glaring since the 2008 financial crisis. Much of the current literature on stimulus packages discusses how to make the policies more effective, but comparatively little has been written empirically on the outcomes of these policies.

Countercyclical policy is considered to have a positive effect on economic recovery among most literature on the subject, with a few notable exceptions. A 2008 report (Spilimbergo et al.) on fiscal policy and worldwide recessions argues that, among the economies with the greatest demand drop resulting from the crisis, a timely, substantial, and targeted stimulus package is the best way to achieve full economic recovery. Drawing upon evidence from prior financial meltdowns like the Great Depression on the 1930s, the East Asian financial crisis of 1997 in Japan and Korea, and the Nordic crisis of 1990-1994, the authors argue that an "early, strong, and carefully thought out fiscal stimulus" is the best way for a nation to succeed in "rescuing the financial sector and increasing demand" (Spilimbergo et al. 2008, 12). Facets of these policies include increased infrastructure maintenance spending, expansion of targeted transfer payments such as unemployment benefits, and temporary reductions in consumption tax. These proposals are also touched on in a 2009 paper about the United States' "road to recovery" after the recession. The paper argues that substantial restructuring and stimulating of the financial sector should be a major part of recovery plans, especially in countries with low imports to GDP ratios. (Zhang and Zhang 2009, 7) Many of the major democracies affected by the crisis, including the United States, France, and Germany, fit this economic profile.

A 2012 paper (Armingeon) similarly argues that while the default reaction to such downturns in a crisis is a mild and only partially effective expansionary fiscal policy, certain countries and governments break this mold with strongly countercyclical policies, causing major economic rebounds in the short run. However, this only tends to occur when there are not "lengthy negotiations or significant compromises between governing parties with different views on economic and fiscal policy" in the course of policymaking (Armingeon 2012, 9). While examples of such situations come primarily from authoritarian systems, de facto one-party democratic systems can and have implemented extended stimulus and recovery programs. This argument assumes that countercyclical policy alone leads to recovery but the importance of a unified government in leading to better recovery deserves further examination.

Another major factor in economic recovery is the role, size, and effectiveness of regulation regimes within the financial sectors of economies. The regulations frameworks of many major economies, particularly those which permitted high levels of public debt as in Europe, were flawed leading up to the 2008 crisis. (Porokowski 2011) Some mechanisms have since been proposed to address these flaws. Porokowski finds that countries that adopted regulatory regimes which monitored risky or overleveraging behavior by banks, such as the Czech Republic, Poland, and Norway, weathered the crisis without major downturns. Moreover, countries that commit to reforming their regulatory regimes without major public debt increases recover more quickly (Porokowski 2012, 194). The role of effective regulation and the scope of regulatory regimes on economic recovery also deserves further examination.

Credit accessibility and availability, an important driver of consumption spending and household debt, was greatly reduced during the 2008 crisis. Not only is credit one of the most important aspects of a growing economy, lackthereof is also a major hinderance to recovering economies. Biggs et al. (2010) sought to explain and debunk the theory of a recovery in which credit does not also re-expand to pre-crisis levels, also known as the "Phoenix Miracle." While prior research compared GDP with the current stock of credit during a single moment of recovery, this study examined the growth and flow of credit instead. The authors found that GDP growth recovery is correlated and temporally associated with a recovery in credit levels (Biggs et al. 2010, 16-17). In other words, credit growth is an important factor to economic recovery.

While the growth of credit during recovery can be positive, in some cases, and certainly in the case of the 2008 financial crisis, it can be highly problematic. According to Jorda et al. (2011), not only are financial crisis recessions simply more costly, but more credit-intensive economic expansion can also lead to a slower recovery. Assessing data on dozens of countries from 1870 to 2008, the authors conclude that a major factor in economic recovery is the extent to which credit played a role in the expansion before the crash (Jorda et al. 2011, 38). This finding is also supported by an IMF report that assessed data from the 2008 crisis, concluding that countries with larger booms in bank credit before the crisis generally experience deeper busts and slower recoveries afterward (Aisen and Franken 2010, 25).

The factors reviewed in the literature of countercyclical policy are innately intertwined. Fiscal stimuli are more likely with unity governments and one-party systems, uncertainty is derived from non-unity governments and regulatory systems being changed, and credit recovery varies based on how dependent the pre-recession economy was on credit. While many factors can potentially affect economic recovery and the efficacy of stimulus packages, there have not been any major post-2008 crisis studies on what the key factors are in determining recovery rates, in part because the global economy has not yet entirely recovered from the recession. This study aims to fill that gap by considering what I believe to be the most important individual factors: stimulus packages, bank leverage, as well as possible confounding factors such as population growth and GDP per capita. This study will examine these factors and attempt to determine how economies can expedite the process of economic recovery.

Theory

Fiscal policy has been a popular and effective tool used by governments to influence economies since John Maynard Keynes popularized the strategy of attempting to counteract and thereby alleviate economic contractions at the macroeconomic level. Especially during the 2008 crisis, countercyclical policies were most visible manifest in stimulus packages worldwide. In the United States, there has been a major debate over the effectiveness of the 2009 stimulus package enacted by President Obama. In order to understand the impact of these policies and advance the debate, it is first necessary to understand the theory behind them.

When an economy crashes due to major shock, such as the bursting of the US housing bubble in 2008, a number of effects often follow. First, banks that have loaned out large sums of money suddenly find that too few entities can afford to repay them. The banks therefore stop giving out loans, causing businesses to lack the credit needed to invest in their own production and growth. At the same time, individuals who have defaulted on their loans have greatly reduced purchasing power. Similarly, struggling and overleveraged banks are unable to grow their portfolios.

Accordingly, overall demand in the economy soon drops significantly. With fewer people buying goods and corporations investing in their production, firms are forced to cut both jobs and output, leading to unemployment and ultimately lowering consumer demand even further. This spiral can continue for years if left unchecked. An economic stimulus package aims to supplement falling consumer demand with increased government demand, jump-starting the economy by injecting a large amount of money through temporarily increased government spending. This continues until all the debt left over from the crash is paid off, banks become solvent once again, and consumer spending rises back to normal levels.

What this concept means is that governments that institute major stimulus packages should have faster recovery and GDP growth rates than those that do not. This effect should be evident for all years after the package is enacted, not just the year that it goes into effect. In addition, the effect of such a stimulus should theoretically be greater than the cost of the stimulus package. While Keynesian economic theories say that it works, the evidence since 2008 must corroborate the story in order to ensure that the theory is sound in this particular case.

Hypothesis and Research Design

The main hypothesis of this paper is that democratic countries that implemented major stimulus packages or countercyclical policies will grow at faster rates on average than countries that did not, while controlling for aggregated bank leverage ratios from 2007, GDP per capita, and population growth.

The independent variable of the hypothesis is whether or not a country enacted a major stimulus package or countercyclical policy after the crisis began. For each country year, the data table specifies whether or not, in the current year or before, there was a major countercyclical fiscal policy initiative. The dependent variable is the growth rate of GDP in each country after the crisis. Major stimulus packages are defined as specific legislation passed to stimulate demand and growth that exceeds 1% of GDP. This figure will reduce interference from routine spending increases while still maintaining a large sample size of stimulus programs. Data on the stimulus packages will be drawn from news media or public releases and OECD statistics on GDP. The annual GDP growth rates will be analyzed, starting from 2009. A democracy is defined as any country rated "Free" by Freedom House in their 2012 Democracy Index. Each country year will have one data point, 1 or 0 respectively, for whether or not it enacted a major stimulus package that year or previously.

To increase the validity of conclusions, this study will control for a number of variables in the statistical regression. The first such variable is bank leverage ratios. The 2007 aggregated bank leverage ratios among the selected cases show the ratio between liquidated assets and loans for each country in that year. In other words, if a bank loans out 5 dollars for every dollar it has in current assets, the leverage ratio of that bank is 5. Once a credit bubble bursts, many of the loans a bank funds do not get repaid. If the leverage ratio is too high, it becomes difficult for a bank to cover the losses on the defaulted loans. The more leveraged banks are, the longer the recovery period needed to return to solid financial ground and begin lending again. Accordingly, the pre-crisis leverage ratios should theoretically be a significant factor in the GDP growth for each country year. The bank leverage ratios are taken from the research of Kalemli-Ozcan et al. (2012). Each country will have the same leverage value for all 4 observation periods, as it is only the pre-crisis leverage peak of 2007 that is expected to have an impact on recovery.

A second control variable is the population growth of each country year, with data from the World Bank online database as well as the CIA's published statistics for 2012. People produce goods and services, and it seems intuitive to say that more people will produce more goods and services. As a country expands, either through childbirth or more importantly immigration, its GDP generally expands to accommodate the increased workforce as well. Because GDP is a measure of all goods and services produced each year, and since a larger population should almost automatically produce more goods and services, varying levels of population growth could have the effect of swaying the data and should be controlled for.

A last control variable is the wealth of a country, operationalized as GDP per capita. Richer countries tend to experience different effects from economic policies, and the concept of economies of scale implies that the larger an economy is, the more effective a package over 1% of GDP would be at instituting growth. Additionally, richer economies that have already achieved technological productivity gains and have higher levels of capital generally have lower growth than poorer countries, as basic increases in technological spread or capital levels have decreasing marginal as a percentage for GDP. Due to both confounding pathways, I will control for wealth in order to get a clearer picture of the effects of stimulus packages across all economies.

Cases and observations

The cases selected are all democracies, as defined by Freedom House, and for whom the relevant data is available from the 2008 financial crisis to the present. This is because democratic countries are the countries with the resources, developed economy, and effective political systems to weigh the option of whether or not to implement a stimulus package. Democracies are also much more likely to have the financial resources to do so, as well as to possess the developed financial sectors impacted by the 2008 financial crisis. Additionally, democracies are much more likely than non-democracies to have accurate, reliable statistics. The study will cover 87 country cases, covering a range of recovery rates, leveraging ratios, uncertainty in economic policy, and stimulus packages. Because bank leverage ratios and uncertainty data is not available for all country years, the number of observations is adjusted accordingly for those two variables. Only country-level data on each of the variables is considered.

Method of Analysis

Due to the large volume of data and cases, quantitative methods are used for analysis. To test the hypothesis, I first use a bivariate regression between major countercyclical policy initiative and GDP growth. I follow with a multivariate regression including the aforementioned control variables to ensure the findings are not skewed by confounding factors.

Data Analysis

For the first regression, analyzing only stimulus and its effect on GDP growth without factoring in control variables, the B value shows that on average, countries that implemented major stimulus packages or countercyclical policies grew about 2.14 percent more per year than those that did not. The correlation is highly significant, with a P value = .005. This corroborates the hypothesis that countercyclical fiscal policy initiatives bolstered recovery rates following the 2008 financial crisis.

The second regression, controlling for bank leverage, population growth, and GDP per capita, further strengthens the validity of the hypothesis. However, the regression also contradicts some of the arguments behind the control variables themselves. The additional variables actually increased the effect of a stimulus package. In this regression, countries that initiated stimulus packages actually grew 2.38% faster on average from 2009 to 2012 than those that did not. The significance of this correlation also improved, with a P value = .003, expressing a confidence level of over 99%.

Some findings merit further consideration. The coefficients on bank leverage ratios are negatively correlated with higher 2007 leverage ratios and GDP growth from 2009-2012. Each unit increase in the leverage ratio corresponded to a .083% drop in GDP growth, a seemingly small correlation. However, given an average leverage ratio of around 14, with some ratios in Germany and the UK going as high as 30, this can account for much of the difference between the first and second regressions' demonstrated effect of countercyclical policy on GDP growth from 2009-2012. However, this correlation is insignificant, with a P value = .168, and therefore does not support the theory behind controlling for bank leverage ratios.

Population growth was more significant and adhered more closely to the hypothesis. Population growth even outperformed my expectations as a factor in GDP growth and economic recovery. Each additional percent of population growth corresponded to a 1.459% increase in GDP growth, with a P value = .004, indicating strong confidence in this correlation. Failing to adjust for population growth accounts for much of the divergence between the first and second regressions and the effect of stimuli on GDP growth.

GDP per capita had a negative impact on growth, with each additional

$1,000 per capita corresponding to a .046 drop in the growth rate. However, GDP per capita was narrowly insignificant, with a P value of .073, just above the .05 significance threshold.

The following model summarizes the impact of countercyclical policy in explaining the variation in GDP recovery rates after 2008.

The adjusted R-Square measure shows that the model can account for about 12.1% of the variation in GDP growth among democracies from 2009 to 2012. Considering the high number of factors that influence growth and the high volatility of growth after the crisis, this R-Squared value does support the validity of the model and thus the validity of the results.

Conclusions

There is a strong, significant, and positive correlation between stimulus package implementation and faster recovery after the 2008 financial crisis. Controlling for population growth, bank leverage rates, and GDP per capita, countercyclical stimulus packages increased growth by about 2.4% annually. Additionally, this model explains around 12% of the variation in post-2008 GDP growth rates. The model itself is clearly not all-encompassing. However, these policies remain highly effective tools to increase GDP growth during economic recoveries and should be considered by policymakers following financial crises. Overall, the hypothesis is supported by the data. Economies that implement stimulus packages grew faster, controlling for other factors, after the financial crisis than those that did not.

Due to the constraints of only accounting for 12% of the variation, the implications of this study are not wide in scope but may advance responses to future financial crises. In the aftermath of the 2008 crisis, there is evidence to conclude that instituting stimulus packages over 1% of GDP is a significant positive factor in recovery. This study can serve to counter critics who claim stimuli categorically cannot improve GDP growth. This is not to imply that all stimulus packages are good, or to imply that countercyclical policy is the "right" choice for any country undergoing a financial crisis. Rather, this study is simply exhibiting empirical evidence that post-2008, countries that implemented such packages grew faster on average.

Nevertheless, this study has considerable limitations. First and foremost, while 1% of GDP seemed to be an effective cutoff for the countercyclical policy size, it is possible that another threshold might be more effective in distinguishing between major stimulus packages and smaller policy initiatives. Another factor the study did not incorporate is the effect of economic interactions between countries. Countries with high levels of trade can impact growth among each other, and a country with especially high or especially low growth can distort the growth of its trading partners. There may also be ways to improve the operationalization of control variables. For instance, the released aggregate bank leverage data may not be completely comprehensive or accurate. There was also not enough data or faith in the theory to include the effects of policy uncertainty, and only having data for certain countries on leverage may have skewed the results.

Future research should focus on additional factors that may affect economic recovery and work to integrate research on each individual factor into an overall study across multiple financial crises. Additionally, certain factors identified in the literature review, such as regulatory regimes or uncertainty, have little to no data available to date. As this data becomes available, it could shed much more light on the effect those factors have on recovery and growth.

Author

Bradley Harmon is a student of International Relations and Political Science, specializing in National Security Policy in East Asia. School of International Studies/School of Public Affairs, American University.

References

Aiginger, Karl. "Why Performance Differed Across Countries in the Recent Crisis." WIFO. http://www.oecd.org/social/labourmarketshumancapitalandinequality/48197530.pdf (accessed November 7, 2012).

Aisen, Ari and Franken, Michael, Bank Credit During the 2008 Financial Crisis: A Cross-Country Comparison (February 2010). IMF Working Papers, Vol. , pp. 1-25, 2010. Available at SSRN: http://ssrn.com/abstract=1562415

Anderson, Robert "Finland Plans 2 Billion Euro Stimulus Package." Financial Times. http://www.ft.com/cms/s/0/ba1d488a-eef0-11dd-bbb5-0000779fd2ac.html#axzz2BYonZJ2N

Armingeon, Klaus "The Politics of Fiscal Responses to the Crisis of 2008-2009." IPW. http://www.ipw.unibe.ch/unibe/wiso/ipw/content/e2425/e2426/e93619/e95082/files95083/KlausArmingeonThepoliticsoffiscalresponsestothecrisisof2008-2009Governance_ger.pdf (accessed November 23, 2012).

Baker, Scott, Nicholas Bloom, and steven Davis. "Measuring Economic Policy Uncertainty." University of Colorado. http://leeds-faculty.colorado.edu/bhagat/PolicyUncertainty-davis.pdf (accessed November 23, 2012).

Biggs, Michael, Mayer, Thomas and Pick, Andreas, Credit and Economic Recovery: Demystifying Phoenix Miracles (March 15, 2010). Available at SSRN: http://ssrn.com/abstract=1595980

Bloom, Nicholas. "Policy Uncertainty Is Threatening Economic Recovery." British Politics and Policy at LSE. http://blogs.lse.ac.uk/politicsandpolicy/2012/04/13/policy-uncertainty-bloom/(accessed November 23, 2012).

"Central Intelligence Agency World Factbook." CIA https://www.cia.gov/library/publications/the-world-factbook/fields/2002.html (accessed November 23, 2012).

Cordero, Jose. "Costa Rica During the Global Recession." Center for Economic and Policy Research. http://www.cepr.net/index.php/publications/reports/costa-rica-during-the-global-recession (accessed November 7, 2012).

"Factbox: Europe's Fiscal Stimulus Plan." Reuters. http://www.reuters.com/article/2009/03/17/eu-summit-stimulus-idUSLH44404720090317 (accessed November 7, 2012).

"G20 Statistical Update: Brazil." International Labor Organization. http://www.dol.gov/ilab/media/events/G20_ministersmeeting/G20-brazil-stats.pdf (accessed November 7, 2012).

"Global Jobs Pact Country Scan: El Salvador." International Labor Organization. http://www.ilo.org/wcmsp5/groups/public/---dgreports/---integration/documents/publication/wcms_178011.pdf (accessed November 7, 2012).

Greeley, Brenden "Krugmenistan Vs. Estonia." Businessweek. http://www.businessweek.com/articles/2012-07-19/krugmenistan-vs-dot-estonia (accessed November 7, 2012).

Jorda, Oscar, Moritz Schularick, and Alan Taylor. "When Credit Bites Back: Leverage, Business Cycles, and Crises." National Bureau of Economic Research. http://www.nber.org/papers/w17621 (accessed November 23, 2012).

Kalemli-Ozcan, Sebnem, Bent Sorensen, and Sevcan Yesiltas. "Leverage Across Firms, Banks and Countries." University of Houston. http://www.uh.edu/~bsorense/leverage_feb28d_12.pdf (accessed November 7, 2012).

Lukasz Prorokowski, (2011) "Recovery from the current banking crisis: Insights into costs and effectiveness of response regulations", Qualitative Research in Financial Markets, Vol. 3 Iss: 3, pp.193 - 223

"Newly Released Economic Stimulus Plan Reaffirms Belgium as Innovation-Friendly Country." Ernst and Young. http://tmagazine.ey.com/insights/newly-released-economic-stimulus-plan-reaffirms-belgium-innovation-friendly-country/ (accessed November 7, 2012).

"Oecd Statistics." OECD.StatExtracts. http://stats.oecd.org/ (accessed October 22, 2012).

"Policy Responses to the Economic Crisis: Investing in Innovation for Long-Term Growth." Organization for Economic Cooperation and Development. http://www.oecd.org/science/innovationinsciencetechnologyandindustry/42983414.pdf (accessed November 7, 2012).

Puddington, Arch. "FREEDOM IN THE WORLD 2012: THE ARAB UPRISINGS AND THEIR GLOBAL REPERCUSSIONS." http://www.freedomhouse.org/sites/default/files/FIW 2012 Booklet_0.pdf (accessed October 22, 2012).

Schwab, Klaus. "Global Competitiveness Reports." World Economic Forum. http://reports.weforum.org/global-competitiveness-report-2012-2013/ (accessed November 23, 2012).

Sorensen, Brent, Sebnem Kalemli-Ozcan and Sevcan Yesiltas. "Leverage Across Firms, Banks and Countries." University of Houston Press (2012), http://www.uh.edu/~bsorense/leverage_feb28d_12.pdf (accessed October 22, 2012).

Spilimbergo, Antonio, Symansky, Steven, Blanchard, Olivier J. and Cottarelli, Carlo, Fiscal Policy for the Crisis (February 7, 2009). Available at SSRN: http://ssrn.com/abstract=1339442

Sterger, Gerhard "Budgeting under and Beyond the Stimulus: Lessons Learned." Organization for Economic Cooperation and Development. http://www.oecd.org/gov/budgetingandpublicexpenditures/44308950.pdf (accessed November 7, 2012).

Victoria, Main. "Hungary Plans Huf 1,400 Billion 2-Year Stimulus Package." FX Street. http://www.fxstreet.com/news/forex-news/article.aspx?storyid=9ff8aa90-731b-49dd-ba16-b9237b3d3e83 (accessed November 7, 2012).

"World Bank Database." World Bank. http://data.worldbank.org/ (accessed November 23, 2012).

Zhang, Yanchung, Nina Thelen, and Aparna Rao. "Social Protection in Fiscal Stimulus Packages: Some Evidence." United Nations Development Program. http://web.undp.org/developmentstudies/docs/socialprotection_fiscalstimulus_march2010.pdf (accessed November 7, 2012).

Zhang, Zhiwei, and Wenlang Zheng. "The Road to Recovery: Fiscal Stimulus, Financial Sector Rehabilitation, and Potential Risks Ahead." Science Direct. http://www.sciencedirect.com/science/article/pii/S1049007811000236 (accessed November 23, 2012).