Many commentators have compared the global crisis to the Great Depression. This column explores lessons that can be applied to help shape expectations and guide exit policy for central banks. It argues that the need for credit stimulus should end when failed intermediaries are resolved and positive net present value credits are reallocated to solvent lenders.
During the financial crisis of 2007-8 and the ensuing recession, policymakers undertook a variety of actions aimed at alleviating distress (Eichengreen and O’Rourke). In many ways, US policymaking during the current financial and economic crisis can be viewed as a reaction to the events of the 1930s and as an attempt to avoid a similarly severe recession. Some economists have suggested that, without the plethora of stimulus policies the severity of the crisis would have been significantly worse and similar in magnitude to the Great Depression (Del Negro et al. 2009).
While we will never truly know the counterfactual, we do know that those types of broad stimulus policies – in both the US and abroad – are expensive and weighing heavily on the fiscal burdens of developed countries around the world. As a result, policymakers are keenly aware of the need for a prompt and timely exit from the stimulus programmes that have, so far, been judged favourably.
Where’s the door marked “Exit”?In a recent policy paper, we explore lessons from the Great Depression that be used to help shape expectations and guide exit policy (Mitchener and Mason 2010). In that paper, we develop a definition for policy exit and show that, in reality, exit from Great Depression policies only occurred roughly twenty years after the onset of expansionary policies. Indeed, some popular expansionary institutions that played a part in today’s crisis, the government sponsored enterprises, for one – are holdovers from expansionary
Depression policies. Hence, exit is important, but it is slow.
In order to characterise exit it helps to be clear about what is being foregone. In the simplest sense, most central banks have an explicit or implicit long-run inflation target as well as a target rate of growth for the economy’s output (usually equal to potential GDP). Expansionary monetary policy, therefore, can be characterised as that which targets above-average growth rates in order to foster recovery. Eventually central banks will have to once again target “normal” growth rates, forgoing the extra “juice” meant to facilitate recovery.
Of course, that extra monetary policy stimulus is only optimal within a short-run policy horizon. We argue that the signal for an exit strategy is the decision to stop policies that emphasise the short-run over the long-run. We therefore define an exit strategy as a movement back to policies associated with steady-state growth.
In a typical recession, an exit constitutes a strategy aimed at returning to the medium and long-run targets for inflation and growth, perhaps embodied in central bank reaction function like the Taylor Rule. Exiting from recessions associated with credit crises are more difficult, however, because of the breadth of policies invoked to combat the crisis. Exiting a credit crisis, therefore, involves political and central bank compromises to reduce or eliminate wide-ranging support programmes, a process far more complex than mere monetary policy tightening. Such policies – while offering clear short-run gains – can create substantial difficulties for long-run growth prospects when skewed market allocations and government subsidies deter investment and growth.
The three R’s: Resources, reallocation, and resolutionWe characterise the emergency support in response to financial crises as the “three R’s”:
- provide resources for growth;
- reallocate of financial industry assets; and
- resolve failed institutions.
A key theme that emerges from our analysis is that there is a troubling lack of knowledge regarding bank behaviour both in the Great Depression and today. In both periods, it is not fully understood how to get financial institutions to lend again after a crisis. In both periods, banks built up “excess reserves” and reduced their lending. It may be optimal for banks to do this in the wake of a financial crisis (too few positive net present value projects to invest in, a desire to recapitalise at the equivalent or higher level relative to assets, or heightened risk aversion), but it complicates the central bank’s decision to remove itself from credit markets and return banking activities to markets. We suggest that the Fed would have a better sense of what to do today if we knew more about why banks failed to lend in the 1930s, why they held excess reserves, and what the potential effects of excess reserves are on price dynamics.
Lost independence, political barriers, and the need for a policy metricWe make three observations about exit strategies based on our definition and our research on the Great Depression.
First, stimulus can turn out to be problematic if it leads central banks to veer too far away from independence and toward government credit policy. An exit strategy requires judgment and quite often the decision to wind down or liquidate institutions that were created to combat a crisis; however, those decisions can become politicised, and in some cases, this had led to the use of short-term policies long past their shelf life. As a result of the creation of programmes to support credit and financial markets, central banks can evolve into organs that allocate credit for political reasons. One casualty can therefore be central bank independence.
The Federal Reserve worked together with the US Treasury to maintain low borrowing rates through the Great Depression and WWII. As a result, the Fed’s balance sheet grew enormously during World War II. Moreover, true Fed independence was not obtained until the Fed-Treasury Accord of 1950, which allowed the Federal Reserve to target interest rates without concern for Treasury borrowing costs as it had done almost since its inception. Hence, monetary policy “exit” from the Great Depression really only occurred in 1950.
Second, the political economy of exit for non-central bank programmes will likely be even more challenging. Agencies created during a crisis may are often supported by champions who stand to gain electorally or personally. The Reconstruction Finance Corporation is an example of an agency that served short-term benefits in recapitalising banks and other firms, but continued to credit to private firms into the 1950s, again long after the distress of the 1930s had ended. For both Central Bank and government policy programmes, therefore, exit from the Great Depression did not occur in the 1930s – it took until the 1950s.
Third, since exit can be delayed for political economy reasons, we suggest a metric for timing the shift from short-run to long-run objectives – one that blends the dual objectives of monetary and bank policy. Having a policy target that relates directly to the distress that necessitated the emergency expansionary policies can help guide long-term decision making and defend seemingly costly (in the short-run) policy reversals to steady state norms.
How to keep it organicIn terms of the “Three R’s” taxonomy presented above, exit conditions for resolution and reallocation are largely organic: once failed banks are resolved and loans get reallocated, the work is done. That doesn’t mean the process happens quickly. It takes some banks twenty years or more to be resolved. It may nevertheless be possible to use economic information on the state of the resolution process to guide the exit strategy.
We propose that bank and non-bank resolutions can be used to parameterise less quantifiable monetary and bank policies when traditional monetary targets are rendered inoperable or highly suspect as the result of extraordinary monetary policy measures to combat a financial crisis (zero interest rate policies, quantitative easing policies, etc.). That is, the need for credit stimulus should end when failed intermediaries are resolved and positive net present value credits are reallocated to solvent lenders.