Guest post: Market Monetarism and Financial Crisis
by Vaidas Urba
Market monetarists agree that stable NGDP path is the policy goal. They usually go on to argue that monetary policy should have a single target: market expectations of NGDP, and a single instrument: the size of the monetary base. Often the EMH is considered to be true, and accordingly, it is not important what assets do central banks hold. Sometimes it is even argued that the central bank is not a bank.
Contrast this to the approach found in Woodford’s May 2009 presentation to the Bank of England, where there is an additional instrument of credit easing that is used to address large financial disruptions. In this post I will try to give this second instrument a distinctly market monetarist flavor.
Economic agents are concerned with the monetary environment in two ways – they want to know the forecast of NGDP, and they want to know the forecast of NGDP volatility. The best way to forecast the NGDP would be to use NGDP futures market data, and the best way to forecast the volatility of NGDP would be to look at NGDP options market (unfortunately these markets do not exist yet). While central banks are in a full control of NGDP expectations, according to the traditional market monetarist perspective they can do nothing to influence the expected volatility of NGDP (apart from implementing NGDPLT regime itself).
If we introduce financial market frictions, and examine what happens when markets are not very efficient, we will notice that by selling and purchasing NGDP options, central banks can move the market price of NGDP options, thus changing the ex-ante market estimate of NGDP volatility. When central banks reduce the price of NGDP options, they decrease the probability of a financial crisis, risky portfolios become more attractive, and ex-post volatility of NGDP is reduced. During normal times, central banks have little power to influence NGDP option markets (although a case could be made for a countercyclical NGDP option intervention in order to lean against asset market bubbles while keeping NGDP futures peg unchanged). During financial panics, when market efficiency takes a big hit, the power to change the market estimates of NGDP volatility becomes more significant, and by selling NGDP volatility insurance to the market, central banks can perform the lender of the last resort function in a transparent and non-discriminatory manner while avoiding bailouts and subsidies.
So in this version of market monetarism there are two monetary targets – expected NGDP and expected volatility of NGDP, and two instruments – size of the monetary base and interventions in the NGDP option market. The second target and the second instrument are less important than the first, as the ability to influence the NGDP options market is weak. In addition, it is not possible to avoid discretion when setting the NGDP volatility targets and determining the size of NGDP option interventions.
Purchases of risky assets by the central banks can be justified on market monetarist grounds to the extent these assets replicate the NGDP options. There is a clear difference between the asset and liability operations of the central bank – you issue base money to increase expected NGDP, you purchase risky assets to reduce the expected volatility of NGDP. Selgin’s asset purchase proposal and ECB’s LTRO collateral framework operate with a wide range of financial markets, so their effects are likely to be similar to the sale of NGDP insurance. We can see that the equity capital of central banks is important, without the capital cushion, a central bank would not be able to credibly intervene in the NGDP option markets. This is why central bankers are very concerned with the strength of their balance sheet; they know that macroeconomic volatility would be very high if central bank is undercapitalized.
Is the fiscal cliff dangerous? Yes. Sudden changes in fiscal environment increase the value of NGDP options. Even if NGDP expectations do not change due to reaction by monetary authorities, it becomes likelier that NGDP will overshoot or undershoot the target path temporarily, as it is very easy to make mistakes when forecasting an impact of a large fiscal change. So market monetarists should argue for a gradual fiscal consolidation, even when NGDP level targeting regime is in operation. Of course, the fiscal cliff is especially dangerous when monetary policy does not have a level targeting feature.
We can use this framework of dual monetary policy targeting to arrive to a better understanding of Fed’s policy during the Great Recession. However, instead of NGDP and its volatility, inflation and its volatility will be used here, as it more closely corresponds to Fed’s thinking. Fed Governor Frederic Mishkin has explained it in March 2008: “Although a distinctly different concept from inflation expectations, policymakers need to be concerned about any widening of inflation uncertainty. Indeed, an increase in inflation uncertainty would likely complicate decision making by consumers and businesses concerning plans for spending, savings, and investment.”
In 2007 and 2008, in addition to steering the inflation expectations via the fed funds rate, the Fed has started various asset-side programs (TAF, TSLF, CPPF, and others) that increased its risky asset holdings. In effect, the Fed has sold macroeconomic risk insurance, and by doing this, it has reduced the expected volatility of inflation. In April 2008, Bernanke worried that the size of Fed’s balance sheet would not be sufficiently large to accommodate the need for asset purchases, and there were discussions about starting the payment of interest on reserves. By paying interest on reserves, the Fed would be able to expand the asset-side programs without losing the ability to use fed funds rate to target inflation expectations.
The summer of 2008 was a period when the Fed has started to make serious mistakes about inflation expectations. It is remarkable how all the efforts to reduce inflation uncertainty via asset-side programs came to nothing when deflationary mistakes in the course of the regular monetary policy caused a renewed episode of financial instability that culminated with the default of Lehman Brothers.
After Lehman, the Fed continued its mistakes in steering inflation expectations, in addition, it underestimated the size of asset side operations that were needed to control inflation volatility.
Things got so bad that the Fed had temporarily lost the power to set the fed funds rate, as evidenced by the TED spread and the standard deviation of effective fed funds rate.
In October 2008, the loss of the fed funds rate instrument meant that the inflation expectations were crashing as the misreading of the economic situation by FOMC was compounded by interest rates that were higher than FOMC intended. A large dose of additional asset-side purchases was required until the Fed regained the ability to steer its preferred policy target. The ECB was the most successful of main central banks in adapting to the post-Lehman situation, on 8 October 2008 the ECB announced the switch to the full allotment procedure that on 13 October 2008 was extended to the provision of dollar liquidity. The ECB’s promise of unlimited dollar liquidity against a large pool of diverse European assets was one of the most important actions that led to the restoration of Fed’s control over the dollar fed funds rate, it also contributed to the reduction of inflation uncertainty.
Market monetarism is the best way to fix the liability policy of central banks. NGDP level targeting would have prevented the Great Recession. However, the Great Recession shows us how a financial crisis can disrupt the market that the central bank uses as its instrument. If a financial crisis happens when NGDP level targeting regime is in operation, central banks should be ready to use asset side policy to preserve the integrity of NGDP futures markets. Even when NGDP futures peg is in place, macroeconomic environment can be unstable if expected volatility of NGDP is too high.