چرخه کسب و کار سیاسی آلمان: تقاضای پول به جای سیاست های پولی

Arguably, Germany had the world's most independent central bank. Surprisingly, however, recent work has found political business cycles in German monetary aggregates. It is hard to explain this with standard models of opportunistic government behavior. Instead, we show that the cycles originate from shifts in money demand tolerated by the Bundesbank. Such shifts occur because, when inflation preferences differ between political parties and election results are uncertain, rational investors avoid entering into long-term financial contracts before elections. Contrary to the Bundesbank's stated commitment to a monetaristic policy rule, it appears to have allowed these changes to have an impact on monetary aggregates.

There is a paradox in the literature on central banking. While it is generally believed that central bank independence is a means of preventing opportunistic Nordhaus-type monetary policy around elections, there is strong evidence that the German Bundesbank—repeatedly described as “one of the most independent central banks of the world” Eijffinger and de Haan, 1996 and Cukierman, 1998 and the European Central Bank's role model—tolerated a political business cycle in German monetary aggregates.1 Results to that end were reported by Alesina et al. (1992) and confirmed by Berger and Woitek (1997a), who found an increase in the growth rates of various German monetary aggregates before elections and a matching decline afterwards. The cycle was small in quantitative terms (±0.2–0.4 percentage points) but extremely strong statistically. The result is robust with respect to the specific (real or nominal) monetary aggregate analyzed, the univariate or multivariate specification of the estimated time series model, alternative data frequencies, and different methods of modeling seasonality.2 Including or excluding the turmoils of German unification after 1989 in the analysis also does not change the outcome. There are two possible explanations for the discrepancy between the Bundesbank's reputation and a political business cycle in German money. One is that it is not a contradiction in the first place, but rather a consequence of the interaction between the Bundesbank and the German government. If the Bundesbank Council had partisan preferences, it might in fact (mis-) use its independence to either support or oppose an incumbent government depending on the ideological beliefs of the Council's median voter Vaubel, 1997a and Sieg, 1997. If the partisan beliefs of the two actors coincide (do not coincide) before an election, the bank follows an expansionary (a contractionary) monetary policy stance, which it will then correct after the event. In other words, the opportunistic cycle might be an anomaly in the data. Obviously, this hypothesis can only be tested if the partisan preferences of the Bundesbank can (at least ex post) be known. Assuming the party preferences of a Council member to be identical with those of the government body that nominated the individual, Vaubel, 1993 and Vaubel, 1997a was unable to refute the hypothesis. His results were questioned by Berger and Woitek (1997b) using data on the individual voting behavior of the Bundesbank Council.3 There is an alternative explanation for the coexistence of central bank independence and political business cycles in German monetary aggregates. If the Bundesbank were as devoted a supporter of monetarist principles as it claimed to have been since the mid-1970s, the conclusion that it actively steered monetary policy in an opportunistic fashion around elections could hardly be avoided. On the other hand, the Bundesbank might instead have followed an interest-rate policy aimed at stabilizing the economy around presumed long-run equilibrium values of key variables in its target function such as inflation and real activity Taylor, 1993a and Taylor, 1993b. A number of recent econometric results point in the latter direction Bernanke and Mihov, 1996, Clarida and Gertler, 1997 and Clarida et al., 1998. Section 2 will extend the literature by providing results based on the Bundesbank's performance since the early 1950s. If the above outcomes can be supported, the Bundesbank's money supply was elastic and shifts in money demand rather than money supply determined the volume of monetary aggregates. But in that case, why should money demand have been influenced by politics? The answer may lie in the uncertainty created by upcoming elections. Alesina and Rosenthal (1995) argue that politicians have partisan preferences that are distinct enough to introduce permanent differences in the way they conduct fiscal policy or try to influence monetary policy.4 If this is correct (or at least cannot be ruled out), interest rate forecasts extending into post-election periods depend on a weighted average of the inflation rates expected for all possible election results. To avoid this uncertainty, financial investors can find it beneficial to postpone certain commitments. They will trade longer term assets for shorter term assets and thus—with an elastic money supply—enlarge monetary aggregates as defined, for instance, in M3, M2 or M1.5 If the policy stance of the new government is known with certainty after the election, the process will be reversed at this time.6 For an uninformed outside observer, however, the phenomenon, which is really a demand-induced pattern caused by the uncertainty connected with elections and, possibly, partisan politics, might look like a Nordhaus or Rogoff/Sibert type opportunistic political business cycle. Section 3 provides tests for the impact of elections on money demand. Section 4 contains our conclusions.

There seems to be a paradox in the empirical literature on the German Bundesbank. On the one hand, Germany is often said to have one of the most independent and also one of the most successful central banks in the world. On the other hand, recent work on political business cycles shows that such a cycle exists in German monetary aggregates. One possible explanation for this contradiction is that politics rather than economics actually drove the Bundesbank. The bank might have misused its independence in order to support governments before elections. It is hard, however, to bring this hypothesis in line with the available evidence. The Bundesbank seemed to sometimes postpone interest rate rises until immediately after elections, but it did not cause the political business cycle in German monetary aggregates. We argue (and show empirically) that the answer to the puzzle may lie in the uncertainty created by upcoming elections. If partisan preferences of governments introduce permanent differences in the conduct of fiscal policy or government pressure on the central bank, interest-rate forecasts extending into post-election periods depend on a weighted average of the inflation rates expected for all possible election results. As a consequence, financial investors trying to avoid this uncertainty trade longer-term assets for shorter-term assets. This, in turn, enlarges monetary aggregates because German money supply is sufficiently elastic. Contrary to the Bundesbank's rhetorical commitment to a monetaristic policy rule, this is indeed the case: empirically, its behavior is best described as an interest rate policy rule that set the short-term interest rate to minimize deviations from equilibrium values of inflation and real growth. After the election, when the preferences of the new government are apparent, the demand process driving the money stock up is reversed. For an outside observer, however, the demand-induced pattern caused by the uncertainty associated with elections and, possibly, partisan politics might look like a opportunistic political business cycle. We conclude that our results offer a solution to the apparent contradiction between the Bundesbank's reputation as having been one of the world's most independent and conservative central banks and the traces of straightforward opportunistic business cycles in German monetary aggregates.