Independent, but Not Indifferent: Partisan Bias at the Federal Reserve

As hard as it is to believe, within two weeks we will (probably…) be discussing the policy consequences of the 2012 US Presidential elections as opposed to forecasting its outcome. With that in mind, we are pleased to welcome the following guest post from University of Michigan political scientist William Clark about the effects of the election outcome on the supposedly most independent of US government institutions, the Federal Reserve.


The Federal Reserve Bank behaves very differently when the White House is occupied by a Democrat than when it is occupied by a Republican.   The Fed Funds Rate, the primary lever used by the Fed to manage the nation’s economy, tends to rise during Democratic administrations and fall during Republican administrations (See Figure.  Note:  Blue=Democratic, Red=Republican).  This pattern may help explain  why Republicans have controlled the White House more frequently than Democrats in recent decades. In a forthcoming paper in Economics & Politics, Vincent Arel-Bundock and I argue that the Fed behaves as if it prefers the election of Republican presidents.   We argue that such a preference is consistent with existing theories about the link between central bank independence and inflation.  Further, differences in Fed behavior cannot be easily explained by the fact that Democratic presidents preside over different macroeconomic conditions than Republicans, because the differences in policy remain after we control for macroeconomic conditions.   We also argue that the Fed’s ability to lean in favor of Republican incumbents is not in spite of, but because of its independence of direct political control.

An examination of movements in the Fed Funds Rate between 1953 and 2008 suggests that as elections draw near, the Fed adopts a looser monetary policy when Republicans control the White House.  In contrast, it adopts a tighter monetary policy when Democrats face re-election.    This suggests that independence allows the Fed to lean against the use of macroeconomic policy for electoral purposes if and only if the incumbent is a Democrat.   In other words, the Fed appears to be a conditional inflation hawk – it tends to deny incumbent demands for looser policy in the period before elections, but only if the incumbent making such demands is a Democrat.

Further evidence that the Fed is a conditional inflation hawk can be found in an analysis of the so-called Taylor Rule.   The Taylor Rule is the work-horse model used by economists to examine the way Fed policy responds to changing economic conditions – specifically, deviations from an inflation goal and a output goal.   If the Fed were an inflation hawk, adjustments to the Fed Funds Rate would be closely tied to changes in the inflation rate.   In contrast, if the Fed cared more about economic growth than inflation, we would see interest rates moving along with changes in output.   We show that when Democratic presidents are seeking re-election, interest rates are tied to the inflation rate, but are unresponsive to changes in output.   In contrast, interest rates are tied to output, but not inflation, when Republican presidents are seeking re-election.

Why would the Fed act like a conditional inflation hawk?  For reasons we explain in detail in the paper, the Fed expects policy, all else equal, to be closer to its most preferred policy when Republicans are in power.  Consequently, it can afford to be an inflation hawk when Democrats are up for re-election, but faces a tension when Republicans face re-election.  It can push for its preferred tight policy or it can take its foot off the brake in the hope that this will help prevent a Democratic victory – which would force the Fed to accept even less desirable outcomes in the future.  Thus, when Republicans are up for re-election the Fed faces a choice – compromise now, or compromise later.  The data suggests it is in the habit of compromising now.

Does it matter? A glimpse at the last 100 years is suggestive.  During the first half of the twentieth century, Republicans controlled the White House less than half of the time.   From the middle of the last century until the most recent financial crisis, the Republicans controlled the White House nearly two-thirds of the time. Standing between these two periods was a historic event: the 1951 Treasury-Fed Accord, which, by removing the Fed’s obligation to monetize Treasury debt, gave the Fed operational independence.  This suggests that it may not be despite its independence that the Fed may be able to weigh in on electoral politics, but because of it.

7 Responses to Independent, but Not Indifferent: Partisan Bias at the Federal Reserve

  1. Rob October 25, 2012 at 12:30 pm #

    So Morgan Warstler was right! (in a sense).

  2. Andrew B. Lee October 25, 2012 at 1:38 pm #

    Correct me if I’m wrong, but this helps explain Larry Bartels’ strange finding in “Unequal Democracy” that Democratic presidents tend to experience strong growth in the beginning of their terms and weak growth at the end, whereas for Republican presidents it is the opposite

  3. Andrew B. Lee October 25, 2012 at 1:39 pm #

    *may* help

  4. JVM October 25, 2012 at 4:35 pm #

    This is massively confounded by the fact that economic expansion has predominated during Democratic presidencies and contraction during Republican presidencies, which would independently be associated with fed tightening/loosening. (source:

    Before I get spammed for partisanship I just want to say I think the conclusion that Democrats are “better” at the economy is massively premature on the basis of that correlation; the point is just that whatever the reason democrat presidencies are confounded with economic growth.

    • JVM October 25, 2012 at 4:38 pm #

      btw I know this comment doesn’t really respond to the finer-grained analysis of the Taylor rule so there could be something there, but those analyses are also less reliable due to small samples.

      • Tyler Healey October 26, 2012 at 10:10 am #

        There’s not much evidence that a high federal funds rate produces less economic growth than a low federal funds rate. For example, Alan Greenspan sharply raised interest rates during Bill Clinton’s first term and this did not prevent Clinton’s reelection.

  5. Frankly Curious October 26, 2012 at 11:38 pm #

    I’m still trying to figure out the whole paper. I naturally agree with this. It makes sense and I’m mad as hell:

    Fed Swings Elections for Republicans

    There are other reasons that Republicans could have done better better getting elected President. I’m fond of the theory that the first two years are the most potent for a President. As a result, Republican policies that don’t help the economy wear off by the time of re-election. Democratic policies that do help the economy also wear off by the time of re-election. Since people vote mainly on economic trends, Democrats do worse.

    But I’m really interested in this theory. Why people aren’t in the streets I can’t say.