Pages

Friday, July 11, 2014

It's Time for the Fed to Move Beyond Inflation Targeting (and Officialy Do what Israel and Australia are Unofficially Doing)

I have a new policy paper that argues inflation targeting has passed its expiration date. Here is the title and abstract:
Inflation Targeting: A Monetary Policy Regime Whose Time Has Come and Gone
Inflation targeting emerged in the early 1990s and soon became the dominant monetary-policy regime. It provided a much-needed nominal anchor that had been missing since the collapse of the Bretton Woods system. Its arrival coincided with a rise in macroeconomic stability for numerous countries, and this led many observers to conclude that it is the best way to do monetary policy. Some studies show, however, that inflation targeting got lucky. It is a monetary regime that has a hard time dealing with large supply shocks, and its arrival occurred during a period when they were small. Since this time, supply shocks have become larger, and inflation targeting has struggled to cope with them. Moreover, the recent crisis suggests it has also has a tough time dealing with large demand shocks, and it may even contribute to financial instability. Inflation targeting, therefore, is not a robust monetary-policy regime, and it needs to be replaced.
In the paper I go through the history of inflation targeting and tie together together several existing critiques of it to show that it no longer is an adequate way to do monetary policy. One critique of it that is topical and should be of interest to readers is my discussion of how inflation targeting can contribute to financial instability. A number of observers, including prominent economists like William White and Lawrence Christiano, Roberto Motto, and Massimo Rostagno, have found that even flexible inflation targeting has a propensity to give rise to the buildup of financial imbalances and exacerbate boom-bust cycles. I draw upon their work and provide empirical evidence supporting their claims. 

I also note the failure of inflation targeting to close the output gap in the United States and the Eurozone as further evidence of its limited ability. And for those who like a more rules-based approach to monetary policy I discuss how the evolution of inflation targeting into flexible inflation targeting opens the door for more judgement calls and opportunities for bad calls as monetary policy is executed in real time.

Since most of the paper covers the problems with inflation targeting--there is a brief section at the end on what should replace it--I thought I would plug here another recent article I coauthored with Ramesh Ponnuru that provides a nice follow up. It shows, based on what happened in recent crisis, what should be the next step in the evolution of monetary policy:
A global economic crisis may be painful, but it can provide some useful lessons. Countries recovered from the great Depression in the order that they exited the gold standard of the time, which is a major reason most economists no longer favor that monetary regime. The turmoil of the last few years has followed a pattern as well: The more a country’s central bank has done to keep nominal spending growing at a steady rate, the better that country has done. This international experience adds to an already-strongtheoretical case for keeping nominal spending—the total amount of money spent in an economy—on a predictable path

[...]

Targeting nominal spending avoids these pitfalls [facing inflation targeting]. If the Fed were trying to keep total dollar spending growing at a 5 percent rate each year, for example, it would not need to loosen or tighten in response to supply shocks. Such shocks would alter only the composition of spending, with positive [negative] ones translating into more [less] goods and services at lower prices. To stay on target, the central bank need respond only to shifts in the demand for money balances. It has to increase the money supply when people are more inclined to hold money balances—when, for example, they are scared of economic trouble and want liquid assets—and decrease the money supply when they are rapidly spending money. To put it another way, the central bank must decrease the money supply when money is circulating quickly and increase it when its turnover or velocity is low. And the more markets expect spending to stay on its target path, the more stable velocity should be in the first place.
We provided several figures that make this point. First, here are the paths of total money spending relative to trend for Israel, Australia, the United States, and the Eurzone (the U.S. trend path is adjusteded to reflect CBO's estimates of full-employment NGDP):


As the above figure shows, Israel and Australia roughly stayed on their trend paths while the United States and the Eurozone did not. These developments can be seen in terms of money supply and money velocity deviations from their trends. In order for total money spending to stay on path, monetary policy has to adjust its stance so that changes in the money supply and money velocity offset each other. As seen below, all of these countries were effectively doing that prior to crisis, but only Israel and Australia continued doing so during the crisis. The grey bars show where the United States and the Eurozone did not do the offsets.


Both Israel and Australia are great success stories of what monetary policy can do even in a severe crisis. Israel's performance, in particular, is instructive as seen here. And the closer a central bank kept its total money spending to its trend path the less unemployment it had during and after the crisis:


So yes, it is time for the Fed to move beyond inflation targeting and officially do what Israel and Australia have unofficially have been doing: stabilizing the growth path of total money spending. This is the same thing as targeting nominal GDP.

PS. Josh Hendrickson has a nice paper that shows an welfare-maximizing central bank could aim to minimize the loss function of money supply and money velocity deviations. If this approach were more commonly used--as oppossed to using a loss function of output and inflation deviations--it would put focus back on what central banks at their core do: stabilize total money spending. And it may have helped avoid the problems seen in the second figure above.

5 comments:

  1. interesting post and paper, and i get the idea.. but operationally i'm not sure how it gets carried out? In both Australia and Israel central bank rates had a direct feedback loop into consumer lending mechanisms and loans directly on household balance sheets..something both the ECB and FED lack.

    ReplyDelete
    Replies
    1. While it's less direct in the US, it should still behave similarly, though with perhaps less direct correlation in time and effect.

      The primary immediate effect is upon market psychology via setting expectations, consistent with the 'Chuck Norris' effect discussed by Lars Christiansen, David, and several other MM theorists.

      The FED doesn't need a direct feedback loop to be effective. With proper expectation settings (earned credibility), the market does all the heavy lifting on the FED's behalf.

      Delete
  2. This proposal looks very classical, as quantitative theory of money (MV = PQ), where interest rates didn't get the importance they now have or the inflation targeting wasn't explicit on that sort of basic manuals. I agree that dealing with money supply, and checking the demand of money is an useful complementary tool of monetary policy (it actually has been in many countries), but to abandon every other tool and keep only this one, may be quite risky and optimistic.

    I mean, assuming a Tinbergen sort of posture where tools should be equal or more thatn the objetives, having only a money supply regulation policy and many objetives, output (gap), unemployment, inflation, is probably over optimistic.

    Perhaps, the problem is not inflation targeting itself but the ways (actions, communications and expectations) that Central Banks try to stay on that target. Though, It is clear that monetary policy after the crisis requires more discussion after a radical (o maybe no so radical) change in goals and means.

    ReplyDelete
  3. This proposal looks very classical, as quantitative theory of money (MV = PQ), where interest rates didn't get the importance they now have or the inflation targeting wasn't explicit on that sort of basic manuals. I agree that dealing with money supply, and checking the demand of money is an useful complementary tool of monetary policy (it actually has been in many countries), but to abandon every other tool and keep only this one, may be quite risky and optimistic.

    I mean, assuming a Tinbergen sort of posture where tools should be equal or more thatn the objetives, having only a money supply regulation policy and many objetives, output (gap), unemployment, inflation, is probably over optimistic.

    Perhaps, the problem is not inflation targeting itself but the ways (actions, communications and expectations) that Central Banks try to stay on that target. Though, It is clear that monetary policy after the crisis requires more discussion after a radical (o maybe no so radical) change in goals and means.

    ReplyDelete