See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/228305592 Is There a Political Dimension on Business Cycle Synchronization? Article in Kyklos · August 2011 DOI: 10.1111/j.1467-6435.2011.00509.x CITATIONS READS 3 42 2 authors: Pedro André Cerqueira Rodrigo Martins 23 PUBLICATIONS 88 CITATIONS 12 PUBLICATIONS 72 CITATIONS University of Coimbra SEE PROFILE University of Coimbra SEE PROFILE All content following this page was uploaded by Rodrigo Martins on 21 December 2016. The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately. KYKLOS, Vol. 64 – August 2011 – No. 3, 329–341 Is There a Political Dimension on Business Cycle Synchronization? Pedro André Cerqueira and Rodrigo Martins I. INTRODUCTION There is a long-standing interest in understanding which variables can explain why some countries have higher business cycle synchronization than others. Until now different channels have been proposed with mixed results. Some of the leading candidates are: trade, industrial structure and/or specialization and financial integration. However for most of these channels we are far from reaching a consensus. Regarding the trade channel Frankel and Rose (1998), Imbs (2004, 2006), Calderon et al. (2007) and Baxter and Koupiratsas (2005), provide evidence that countries with higher levels of bilateral trade are more synchronous. However, Fidrmuc (2004) does not find any correlation between business cycles convergence and bilateral trade, arguing that this result is related to intraindustrial trade. As for industrial structure and specialization Imbs (2004, 2006) finds that similar countries are more synchronous, however for Baxter and Koupiratsas (2005) this hypothesis is not robust. Concerning international financial integration, Imbs (2004, 2006) argues that more financially integrated countries are more synchronous and Kose et al. (2003) find evidence that financially open developing economies have their cycles synchronized with the G7 countries. Nevertheless, Heathcote and Perri (2003, 2004) found that the decrease in business cycle synchronization of the US with the rest of the world could only be replicated by a model when international diversification of portfolios increases. Besides these three channels other variables have been used with different degrees of success, such as: belonging to a currency union or a trade agreement, distance  Cerqueira: Faculty of Economics and GEMF, University of Coimbra, Contact: pacerq@fe.uc.pt. Martins: Faculty of Economics and GEMF, University of Coimbra, Contact: rodrigom@fe.uc.pt. We are grateful to Elias Soukiazis and to the participants in the Annual Meeting of the European Public Choice Society (Jena, March 2008) and the European Economic and Financial Society Conference (Prague, May 2008) for useful comments. The usual disclaimer applies. r 2011 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA 329 PEDRO ANDRÉ CERQUEIRA/RODRIGO MARTINS between two countries, degree of development, similarity of exports and imports, etc.1. However, the literature on political cycles shows that economic growth (see for instance Alesina et al. (1997)) may be influenced by electoral and ideological factors. Thus the previous literature of GDP synchronization may be missing a channel that reflects this dimension. This paper purposes to fill in this gap using a panel data framework to check if there is a political channel to business cycle synchronization. The structure of the paper is the following: section two gives a short insight on political cycles theories, and puts forward the hypothesis of how the political variables may interact with business cycle synchronization; the following section describes the data details and the synchronization index used. The fourth section presents and discusses the results and the final one concludes. II. POLITICAL CHANNEL OF BUSINESS CYCLE SYNCHRONIZATION Akerman (1947) and Frey and Lau (1968) are among the first to study the role of electoral considerations on economic cycles. Since then a significant number of papers seek to answer whether electoral motives and government ideology influence short term economic performance. The literature evolved into two main theories, the political business cycle and the partisan theories, and also to a branch that tried to combine the features of both (Frey and Schneider (1978)). Overall, the research on political cycles consistently finds evidence of opportunistic or partisan effects in the macroeconomic environment. For encompassing surveys of the literature see, for instance, Schneider and Frey (1988), Alesina et al. (1997), Franzese (2002) and Franzese and Jusko (2006). If national growth exhibits political effects then, theoretically, growth synchronization between nations may well be explained by them2. This means that the literature on business cycle synchronization may be disregarding the existence of a political channel. The political business cycle theories hypothesise that all politicians try to create similar favorable economic conditions before elections in order to gain voters support. With the seminal work of Nordhaus (1975) a first generation of models relying on adaptative expectations emerged (see also Lindbeck (1976) and Ben-Porath (1975)), and in the late 1980’s and 1990’s a second wave approached opportunistic models in a rational expectations perspective (Rogoff and Sibert (1988); Rogoff (1990)). Empirically, Grier (2008) finds evidence 1. Other related studies are those by Clark and van Wincoop (2001), Otto et al. (2001), Rose and Engel (2002) and Darvas et al. (2005). 2. Sapir and Sekkat (2002) found that German politics impacts the economy of other European countries. 330 r 2011 Blackwell Publishing Ltd. IS THERE A POLITICAL DIMENSION ON BUSINESS CYCLE SYNCHRONIZATION? of an electoral cycle when examining GDP growth in the United States, however these theories seem to receive more empirical support in developing than in developed countries (see Brender and Drazen (2009); Vergne (2009); and Shi and Svensson (2006)). The expected consequences of opportunistic behaviour to business cycle synchronization depend on electoral synchrony. If two countries hold elections at the same time their economies experience similar politically induced economic fluctuations that increase cross country business cycle synchronization, but without electoral synchrony these fluctuations occur out of phase thus decreasing cycle coordination. This paper also tests the linkage between political instability and business cycle synchronization. Alesina et al. (1996) show that GDP growth is significantly lower in countries and in time periods that exhibit a high propensity for government collapse. In a recent paper Jonga-Pin (2009) also finds that higher degrees of political instability tend to lower economic growth. If this scenario occurs in two countries simultaneously then business cycle synchronization may increase whereas if only one country is experiencing political instability GDP synchronization is expected to decrease. The traditional partisan theory (Hibbs (1977)) and the rational partisan models (Alesina (1987); Alesina and Sachs (1988)) assume that different parties have different policy objectives that reflect the preferences of their partisans. When in office left-wing parties are relatively more concerned with unemployment (growth) than with inflation whereas right-wing parties are especially worried with inflation control. Hibbs (1987, 1992) and Alesina et al. (1997) find partisan outcome cycles in OECD democracies. The partisan theory focuses on the role of ideology therefore countries with governments that share the same policy preferences are expected to generate similar economic outcomes. Comparative studies show that left wing governments worsen nominal and improve real and distributional outcomes. As business cycle synchronization deals with growth alignment the simultaneity of left wing governments should increase cross-country correlation. In the case of right wing governments as they are more concerned with inflation one should expect an increase in cross-country alignment of inflation. Regarding growth synchronization, the effect may depend on how each economy reacts to the anti-inflationary measures taken. III. DATA AND MEASUREMENT The analysis of the political channel requires a synchronization indicator at all points in time in order to estimate the following equation: rij;t ¼ FðXij;t ; uij;t Þ r 2011 Blackwell Publishing Ltd. ð1Þ 331 PEDRO ANDRÉ CERQUEIRA/RODRIGO MARTINS where at time t, rij, t represents the GDP cross correlation index between country i and j, Xij, t is the matrix of explanatory variables and uij, t are the idiosyncratic shocks. However previous studies usually resort to a cross-section specification such as: T T X Xij;t X uij;t rij ¼ F ; T T t¼1 t¼1 ! ð2Þ removing the time variability by averaging equation (1). To overcome this problem we use a period by period correlation index based on the distance between standardized variables proposed by Cerqueira and Martins (2009) of the two countries at any given date: 0 rij;t 12 1 B   B CC B CC dj;t  d j 1B ðd  d Þ B i;t i CC s s ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi ¼ B1  B  CC B C 2B T  T    @ P P 2 2A A @ 1 1 d d d d   0 T i;t t¼1 i T j;t ð3Þ j t¼1 where di, t and dj, t are the GDP growth rates of countries i and j. Contrary to the correlation index computed over the entire period, this index enables us to distinguish negative correlations related with specific years, asynchronous behavior over unstable periods and synchronous behavior in a stable economic environment. Although rolling windows is an alternative way to capture the time variability, this correlation index has some advantages. First there is no need to set a window span, second there is no loss of observations, third it does not exhibit the so called ‘ghost features’, as the impact of a major shock is not reflected in n consecutive periods, with n being the window span. Moreover, the major disadvantage with the use of overlapping windows is that the resulting variables are heavily autocorrelated and, thus, difficult to handle in econometric analysis. The index was computed using GDP data from the OECD database 2006 for twenty OECD countries3 from 1970 to 2002. Concerning the economic explanatory variables we account for the most commonly and robust channels found in the literature: trade, specialization and financial openness. 3. The countries are Australia, Austria, Canada, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Japan, the Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, the UK and the USA. Data details are available from the authors upon request. 332 r 2011 Blackwell Publishing Ltd. IS THERE A POLITICAL DIMENSION ON BUSINESS CYCLE SYNCHRONIZATION? As to the trade channel we use the bilateral trade intensity index based in the Deardorff (1998) model, given by: Bilateral Tradeij;t ¼ Expij;t þ Impij;t GDPw;t  GDPi;t  GDPj;t 2 ð4Þ where Expij, t are the exports from country i to j, Impij, t are the imports of country i from j and GDPw, t is the world level of GDP4 at time t This measure does not depend on country size as the one proposed by Frankel and Rose (1998). To access the impact of the industrial structure we use a similarity index given by:  n  X  k 6 j ð5Þ sij;t ¼ si;t  skj;t ; i ¼ k¼1 ski;t represents the weight of sector k in the GDP of country i at time t. where When interpreting the results of this industrial similarity measure one should take into account that they vary between a minimum of 0 (complete similarity between countries) and a maximum of 2 (completely different structures). As for the financial channel we use the financial openness average indicator given by:   Ai;t þ Li;t Aj;t þ Lj;t ð6Þ Financial Opennessij;t ¼ þ GDPi;t GDPj;t where Ai, t and Li, t are total assets and liabilities of country i, at time t. The billateral trade and the industrial structural similarity indexes were computed using the GDP, total imports and exports and the value added at the sectorial level5 from the OECD database 2006 and regarding the financial openness indicator we used the Lane and Milesi-Ferretti (2007) data set. In order to control for institutional factors affecting synchronization we add dummy variables for simultaneous membership of the EEC (later EU) or NAFTA. Regarding the political and electoral environment, we establish two groups of dichotomous variables using the data set and classification of Armingeon et al. (2005). The first tries to capture partisan effects in the economy. To get a clear ideological influence, we take in consideration two dummy variables, assuming the value 1 if both countries have left or right wing governments, 4. As a proxy for the world GDP we use the GDP of the OECD. 5. We build the industrial structure similarity index using the following aggregation of the ISIC codes: 0 1 1 – Food and live animals 1 Beverages and Tobacco; 2 1 4 – Crude materials, except fuels, animal and vegetal oils, fats and waxes 1 Animal and Vegetal oils, fats and waxes; 3 – Mineral fuels; 5 1 6 1 8 1 9 – Chemical and related 1 Manufactured goods 1 Miscellaneous manufactured 1 Commodities and transactions articles and 7 – Machinery and transport equipment. We used this code aggregation in order to have the maximum observations by country and time span. r 2011 Blackwell Publishing Ltd. 333 PEDRO ANDRÉ CERQUEIRA/RODRIGO MARTINS respectively. The second group tries to capture the impact of national elections and the effect of changes in the cabinet composition of national governments. As there are elections without cabinet changes, cabinet changes without elections and both at the same time we test opportunistic and instability effects jointly. The set of dummy variables used covers the revelant scenarios, distinguishing the simultaneity or not of elections and cabinet changes and if these events occur simultaneously or not in both countries. The decomposition is comprised of three variables: (1) Electoral year in at least one country with government changes; (2) Electoral year without government changes; (3) changes in government in at least one country without elections. We also introduce aditional variables in order to test if the simultaneity of political events is significant in each of the three previous scenarios. IV. ESTIMATIONS AND RESULTS This section presents the results. As our model includes autorregressive components we use the Blundell and Bond (1998) GMM methodology with the Windermeijer correction in order to solve the related endogeneity problem and also potential ones concerning economic and political variables. We also use the openness index as instrument for trade intensity6 between countries i and j. When introducing the industrial similarity index we lose a significant amount of observations. For this reason, in addition to the unbalanced panel regressions using the raw data, we estimate the missing values using the multiple imputation algorithm developed by Honaker and King (2010) enabling us to get a balanced panel and to fully exploit the available data. Finally, all economic variables, other than the synchronization index, are in logs. Table 1 presents the two-step GMM estimation results using the GDP synchronization index as the dependent variable. Like other correlations measures, this index may suffer from measurement errors that lead to a larger error variance. But, as it is reasonable to assume that this measurement error and the explanatory variables are orthogonal, the assumptions needed to keep the desirable large-sample proprieties are not altered. Regarding the economic variables, some of the results are in line with previous findings. Bilateral trade has the expected positive sign across regressions showing that more trade between countries increases bilateral 6. We also tried the sum of the GDP per capita of each pair of countries as instrument for trade intensity and industrial specialization n i oand the GDP per capita distance between them, given GDP GDPj by: GDP distanceij;t ¼ Max GDPjt ; GDPit as instrument for industrial specialization, without t t significant changes in our results. Previous studies have used other instruments/variables, such as common border, distance or common language. However these variables are invariant over time and therefore captured by the constant term of the fixed-GMM estimator. 334 r 2011 Blackwell Publishing Ltd. IS THERE A POLITICAL DIMENSION ON BUSINESS CYCLE SYNCHRONIZATION? Table 1 Business Cycle Synchronization Determinants: GMM Panel Estimations: 1970–2002 Unbalanced Panel (1) Constant Bilateral trade intensity Financial openness Industrial similarity (2) Balanced Panel (3) (4) 3.676 2.189 4.017 3.295 (4.13) (4.175) (4.59) (4.33) 0.073 0.0572 0.084 0.073 (1.65) (1.96) (1.91) (2.00) 2 0.670 2 0.363 2 0.780 2 0.657 ( 2 3.70) ( 2 3.37) ( 2 3.70) ( 2 4.20) 2 0.091 2 0.111 ( 2 0.56) ( 2 0.80) 0.098 (1.53) 0.067 (0.59) 0.151 (2.74) 2 0.015 ( 2 0.15) - 2 0.098 ( 2 1.81) 0.129 (2.32) - Election year with government change 0.028 (0.22) Simult. election year with (at least one) 0.156 gov. change (0.44) Simult. gov. change with (at least one) 2 0.116 election ( 2 0.19) Simult. election year; Simult. 2 1.384 gov. change ( 2 1.23) - 2 0.194 ( 2 1.07) 0.256 (0.63) 0.345 (0.52) 2 0.119 ( 2 0.12) - Election year without government change 2 0.246 ( 2 3.30) Simult. election year without 0.074 gov. change (0.32) - 2 0.172 ( 2 2.56) 2 0.249 ( 2 0.85) - 2 0.263 ( 2 3.70) 0.428 (1.42) - 2 0.313 ( 2 2.51) 0.765 (1.63) - EU NAFTA Simultaneous right-wing governments Simultaneous left-wing governments Gov. change without elections Simult. gov. change without elections Observations Arellano-Bond test for AR(1) Arellano-Bond test for AR(2) Groups / Instruments Hansen test 0.116 (1.68) 0.115 (1.06) 2 0.076 ( 2 1.27) 0.100 (1.78) 0.127 (2.70) 0.115 (1.64) 5700 5700 5890 5890 2 9.50 [0.00] 2 9.58 [0.00] 2 9.68 [0.00] 2 9.68 [0.00] 2 0.66 [0.51] 2 0.62 [0.54] 2 0.38 [0.87] 2 0.11 [0.91] 190 / 188 190 / 185 190 / 189 190 /185 161.5 [0.13] 167.1 [0.15] 162.44 [0.14] 167.1 [0.14] ,,- Significant at 10%, 5% and 1% respectively. Numbers in curved brackets are t-ratios and in squared brackets p-values. Regressions include a complete set of time-dummies and lags of the dependent variable (results not shown) synchronization and exibhits point estimates close to the ones found by Frankel and Rose (1998) and Clark and Van Wincoop (2001). We find a negative effect of financial openness, which confirms the results of Heathcote and Perri (2003, 2004). Our results support the international risk r 2011 Blackwell Publishing Ltd. 335 PEDRO ANDRÉ CERQUEIRA/RODRIGO MARTINS sharing theory which states that agents from financially integrated countries can reduce their income risk wether by diversifying their portfolio across countries or through the international credit markets. Either way, if countries are able to reduce their income risk they become free to specialize in sectors that exhibit comparative advantages. With these different specialization patterns weaker synchronization is expected. However, once endogeneity is controlled, we find no significant evidence relating industrial similarity with cross-country synchronization. This result indicates that the transmission channel from financial integration to synchronization might not be due to industrial specialization. A more satisfactory explanation, would be that, if countries are more financially integrated, investment flows faster between them as agents try to exploit short and medium term advantages. These flows of capital, as predicted by the finance literature, benefit countries, that due to transitory idiosyncratic shocks, exhibit higher capital returns, boosting even more their economic performance at the expense of investment and growth in other countries. This increases the investment volatility relatively to the one of GDP and reduces cross-country synchronization. A final point regarding financial openness is why ours and Heathcote and Perri’s (2003, 2004) results differ from those of Imbs (2004, 2006). As Imbs work highlights the cross-section dimension the positive effect of financial openness can be seen as a proxy for long-run institutional factors favouring both financial integration and real synchronization. Differently, Heathcote and Perri’s framework focuses on the time-dimension, while we use a panel where the long-run factors are captured by the fixed effects in the GMM regressions. Therefore, it seems that alternative approaches unveil different effects of financial integration on the business cycle synchronization. The use of a synchronization index suitable for panel data enables us to control for economic integration areas more accurately as we can pinpoint the countries year of entrance. Only EU simultaneous membership has a positive effect on cross-country synchronization, which can be due to increased trade or to higher similarity of the legal economic environment. Looking at the political variables, that as far as we know have not yet been taken into account in the literature, reveals interesting results7. Concerning the ruling party’s (or parties) ideological affiliation we find that simultaneous left-wing governments increase business cycle synchronization. This result is congruent with the partisan theory, because as growth is the priority for these governments it is expected that both countries strive to achieve higher output levels, thus increasing growth alignment. 7. The GMM methodology also controls for potential endogeneity bias in political variables, such as the possibility that governments are more likely to be voted out of office when the economic performance is poor. 336 r 2011 Blackwell Publishing Ltd. IS THERE A POLITICAL DIMENSION ON BUSINESS CYCLE SYNCHRONIZATION? For simultaneous right-wing governments, although the results are statistically weak, they suggest a possible detrimental effect on business cycle synchronization. As stated by the partisan theory, conservative parties use monetary and fiscal policy with the prime objective of controlling inflation levels. So when countries face inflationary shocks, they will control inflation at the short run cost of increased unemployment and lower GDP as stated by the New Keynesian Phillips curve8. However, as time elapses, economic actors can agree to new price and wage contracts expecting lower inflation, so growth and employment return to their natural rates, while inflation remains low. Therefore, when inflationary shocks have asymmetric effects across countries the anti-inflationary measures taken by different right-wing governments will differ leading to different adjustment processes of growth and unemployment. Even with symmetric effects the adjustment rates across countries to the natural equilibrium will be different because economies have different levels of price/ wage flexibility9. In either case with simultaneous right-wing governments business cycle synchronization through growth will decrease (while inflation synchronization is expected to increase) when comparing with a left-right scenario. In this latter case the variability in growth rates is not has high because only one country takes strong anti-inflationary measures. As to the effect of the remaining political variables, the results show that election years with government changes do not affect bussines cycle syncronization, not even when we account for country simultaneity. However, election years without government changes is found to be statistically significant and negatively affecting the business cycle syncronization index, but it seems to be irrelevant whether elections occour simultaneously in both contries or not. One possible explanation for this result is the existence of opportunism, although the evidence found seems to be partial, as business cycle syncronization only decreases across countries when elections do not imply government changes. These results appear to be in line with the hypothesis of Frey and Schneider (1978) that reelection chances dictates whether incumbents behave opportunistically or idelogically. It is possible that governments with litle chances of winning elections chose to act ideologically rather than oportunisically near elections (an effect accounted for by the idelological variables), while governments with a figthing chance at the ballots chose to expand the economy before elections in order to gain voters support, thus decreasing cycle alignment. Furthermore, as the results show a statistically significant effect related only to governments that manage to remain in power, the ability to create the best 8. For details about the New Keynesian Phillips curve see Gali and Gertler (1999) or Sbordone (2002). 9. This asymmetry can be due to differences in the exposure to global price shocks as a result of differences in the commodity intensity of production and consumption or differences in the propagation of shocks due to differences in inflation and wage dynamics. See for instance Vogel et al. (2009) or Galesi and Lombardi (2009). r 2011 Blackwell Publishing Ltd. 337 PEDRO ANDRÉ CERQUEIRA/RODRIGO MARTINS favorable economic conditions in election years may be important, which indicates that cycle alignment may decrease with competent oportunistic governments. The empirical results also suggest that government changes not related to elections reduce business cycle synchronization, regardless of being simultaneous or not. This is in line with Alesina et al. (1996) and Jong-a-Pin (2009) which show that unexpected political turmoil causes instability and uncertainty among economic agents disrupting economic growth. Overall, our findings suggest that cylce alignment might be affected by unexpected cabinet changes and unexpected or, at least, uncertain electoral results. Finally, when comparing the results between models with and without the political channel we find no significant changes in the importance of the traditional economic channels. This shows that political factors are a different and autonomous channel, that improves our knowledge of business cycle determinants. V. CONCLUSION This study started with the assumption that integrating political factors in business cycle synchronization models is likely to improve the knowledge of the phenomenon. To investigate this hypothesis, until now unexplored by the literature, we use a synchronization index that gives us the time variability needed to capture these specific effects. The main conclusion to be drawn from the empirical analysis is that the political environment is a relevant channel that works independently without altering the role of traditional economic channels. To control for the economic determinants, we included a set of variables that have been considered relevant in the literature (trade measures, industrial structure and measures of financial integration). The findings generally confirm the previous research, however regarding financial openness they are coherent with the theoretical expectations but not with the standard empirical literature that uses cross-section analysis. The negative sign found gives us a different perspective of the financial mechanism because one of the main features of the proposed index is the capture of time variability. To evaluate the importance of the political environment we used two groups of variables. The first, measuring ideological effects, was found to be relevant in explaining business cycle synchronization and in line with the partisan hypothesis. Simultaneous left-wing governments improve business cycle synchronization while the results for right-wing governments, although statistically weak, suggest the opposite effect. Our explanation is based on the consensus that left-wing parties are more concerned with growth than rightwing parties. The second group tried to capture the effect of elections and a 338 r 2011 Blackwell Publishing Ltd. IS THERE A POLITICAL DIMENSION ON BUSINESS CYCLE SYNCHRONIZATION? broader spectrum of governmental instability. Although the evidence found seems to be partial, both election years and cabinet changes have a negative impact on business cycle synchronization in certain conditions. Opportunistic behaviour and economic instability due to unexpected cabinet changes may explain this effect on cycle alignment, but the results do not reveal differences when either of these events ocour simultaneously or not in both countries. REFERENCES Akerman, J. (1947). Political Economic Cycles, Kyklos. 1: 107–117. Alesina, A. (1987). Macroeconomic Policy in a Two-Party System as a Repeated Game, Quarterly Journal of Economics. 102(3): 651–678. Alesina, A. and J. Sachs (1988). Political Parties and Business Cycle in the United States, 1948–84, Journal of Money, Credit, and Banking. 20(1): 63–82. Alesina, A., G. Cohen and N. Roubini (1997). Political Cycles and the Macroeconomy. Cambridge Mass. and London: The MIT Press. Alesina, A., S. Ozler, N. Roubini and P. Swagel (1996). Political Instability and Economic Growth, Journal of Economic Growth. 1(2): 189–211. Armingeon, K., P. Leimgruber, M. Beyeler and S. Menegale, et al. (2005). Comparative Political Data Set, 1960–2003. Institute of Political Science: University of Bern. Baxter, M. and M. Koupiratsas (2005). Determinants of Business Cycle Commovement: A Robust Analysis, Journal of Monetary Economics. 52(1): 113–157. Ben-Porath, Y. (1975). The Years of Plenty and the Years of Famine- A political Business Cycle, Kyklos. 28(2): 400–403. Blundell, R. and S. Bond (1998). Initial Conditions and Moments Restrictions in Dynamic Panel Data Models, Journal of Econometrics. 87: 115–143. Brender, A. and A. Drazen (2009). How do Budget Deficits and Economic Growth Affect Reelection Prospects? Evidence from a Large Panel of Countries, American Economic Review. 98: 2203–2220. Cerqueira, P.A. and R. Martins (2009). Measuring the Determinants of Business Cycle Synchronization Usinga Panel Approach, Economic Letters. 102(2): 106–108. Calderon, C., A. Chong and E. Stein (2007). Trade Intensity and Business Cycle Synchronization: Are Developing Countries Any Different?, Journal of International Economics. 71(1): 2–21. Clark, T.E. and E. van Wincoop (2001). Borders and Business Cycles, Journal of International Economics. 55: 59–85. Darvas, Z., A. Rose and G. Szapáry (2005). Fiscal Divergence and Business Cycle Synchronization: Irresponsability is Idiosyncratic. NBER Working Paper 11580. Deardorff, A. (1998). Determinants of Bilateral Trade: Does Gravity Work in a Neoclassical World?, in J. Frankel ed., The Regionalization of the World Economy. Chicago: The University of Chicago Press: 7–32. Fidrmuc, J. (2004). The Endogeneity of the Optimum Currency Area Criteria, Intra-Industry Trade, and EMU Enlargement, Contemporary Economic Policy. 22(1): 1–12. Frankel, J. and A. Rose (1998). The Endogeneity of the Optimum Currency Area Criteria, Economic Journal. 108: 1009–1025. Franzese, R. (2002). Electoral and Partisan Cycles in Economic Policies and Outcomes, Annual Reviews of Political Science. 5: 369–421. Franzese, R. and K.L. Jusko (2006). Political-Economic Cycles, in B.R. Weingast and D.A. Wittman eds., The Oxford Handbook of Political Economy. Oxford: Oxford University Press: 545–564. r 2011 Blackwell Publishing Ltd. 339 PEDRO ANDRÉ CERQUEIRA/RODRIGO MARTINS Frey, B. and L. Lau (1968). Towards a Mathematical Model of Government Behaviour, Zeitschrift fur Nationalokonomie. 28(3/4): 355–380. Frey, B. and F. Schneider (1978). An Empirical Study of Politico-Economic Interactions in the United States, Review of Economic and Statistics. 59: 663–688. Galesi, A. and M.J. Lombardi (2009). External Shocks and International Inflation Linkages a Global VAR Analysis, ECB Working paper series no. 1062. Gali, J. and M. Gertler (1999). Inflation Dynamics: A Structural Econometric Analysis, Journal of Monetary Economics. 44: 195–222. Grier, K. (2008). US Presidential Elections and Real GDP Growth, Public Choice. 135: 337–352. Heathcote, J. and F. Perri (2003). Why Has the U.S. Economy Become Less Correlated with the Rest of the World?, American Economic Review. 93: 63–69. Heathcote, J. and F. Perri (2004). Financial Globalization and Real Regionalization, Journal of Economic Theory. 119: 207–243. Hibbs, D. (1977). Political Parties and Macroeconomic Policy, American Political Science Review. 71: 1467–1487. Hibbs, D. (1987). The Political Economy of Industrial Democracies. Cambridge MA: Harvard Univ. Press. Hibbs, D. (1992). Partisan Theory After Fifteen Years, European Journal of Political Economy. 8: 361–373. Honaker, J. and G. King (2010). What to do about Missing Values in Time Series Cross-Section Data, American Journal of Political Science. 54(2): 561–581. Imbs, J. (2004). Trade, Finance, Specialization and Synchronization, Review of Economics and Statistics. 86(3): 723–734. Imbs, J. (2006). The Real Effects of Financial Integration, Journal of International Economics. 68(2): 296–324. Jong-A-Pin, R. (2009). On the Measurement of Political Instability and its Impact on Economic Growth, European Journal of Political Economy. 25(1): 15–29. Kose, M., E. Prasad and M. Terrones (2003). How Does Globalization Affect the Synchronization of Business Cycles?, American Economic Review. 93: 57–62. Lane, P. and G.M. Milesi-Ferretti (2007). The External Wealth of Nations Mark II: Revised and Extended Estimates of Foreign Assets and Liabilities, 1970–2004, Journal of International Economics. 73(2): 223–250. Lindbeck, A. (1976). Stabilization Policy in Open Economies with Endogenous Politicians, American Economic Review. 66: 1–19. Nordhaus, W.D. (1975). The Political Business Cycle, Review of Economic Studies. XLII(2): 169–190. Otto, G., G. Voss and L. Willard (2001). Understanding OECD Output Correlations, Research Discussion Paper 2001-05, Reserve Bank of Australia. Rogoff, K. (1990). Equilibrium Political Budget Cycles, The American Economic Review. 80(1): 21–36. Rogoff, K. and A. Sibert (1988). Elections and Macroeconomic Policy Cycles, Review of Economic Studies. LV(1): 1–16. Rose, A. and C. Engel (2002). Currency Unions and International Integration, Journal of Money, Credit and Banking. 34: 1067–1089. Sapir, A. and K. Sekkat (2002). Political Cycles, Fiscal Deficits, and Output Spillovers in Europe, Public Choice. 111(1/2): 195–205. Shi, M. and J. Svensson (2006). Political Budget Cycles: Do They Differ Across Countries and Why?, Journal of Public Economics. 90: 1367–1389. Schneider, F. and B. Frey (1988). Politico-Economic Models of Macroeconomic Policy: A Review of the Empirical Evidence, in Thomas D. Willet ed., Political Business Cycles: The Political 340 r 2011 Blackwell Publishing Ltd. IS THERE A POLITICAL DIMENSION ON BUSINESS CYCLE SYNCHRONIZATION? Economy of Money, Inflation and Unemployment. Durham and London: Duke University press: 239–275. Sbordone, A.M. (2002). Prices and Unit Labor Costs: A New Test of Price Stickiness, Journal of Monetary Economics. 49: 265–292. Vergne, C. (2009). Democracy, Elections and Allocation of Public Expenditures in Developing Countries, European Journal of Political Economy. 25: 63–77. Vogel, L., E. Rusticelli, P. Richardson, S. Guichard and C. Gianella (2009). Inflation responses to recent shocks: do G7 countries behave differently?, OECD Economics Department Working Papers. SUMMARY The study of the business cycle synchronization determinants has traditionally focused on economic variables disregarding aspects such as politics and elections. This paper intends to fill in this gap and test whether the political environment is also a relevant channel explaining the synchronization between countries. Using a synchronization index for panel data we find that government ideology affects synchronization. Simultaneous left-wing governments improve business cycle synchronization while the results for right-wing governments, although statistically weak, suggest the opposite effect. In some particular cases, also elections and cabinet changes are found to have a negative impact on synchronization. Furthermore, the role of the traditional economic variables is not altered by the inclusion of this new channel. r 2011 Blackwell Publishing Ltd. View publication stats 341