Why the Fed should raise interest rates at its next meeting

How to Raise Inflation

Conventional New-Keynesian macroeconomists assert that, to increase the inflation rate, the Fed must first lower the interest rate. A lower interest rate, so the argument goes, increases aggregate demand and raises wages and prices. As economic activity picks up, the Fed can raise the interest rate without fear of generating a recession. Some economists advocate that the Fed should raise the interest rate to meet the inflation target, a position that for reasons that escape me, has been labelled as neo-Fisherianism on the blogosphere (see my previous post, also  Stephen Williamson, David Andolfatto, and John Cochrane). My body of work, written over the past several years, (see my forthcoming book) explains how to raise the interest rate without simultaneously triggering a recession and, I suppose, that makes me a 'neo-Fisherian'.

There is also something new in this post that I discussed last week with Stephen Cecchetti while visiting the San Francisco Fed. This relates to the connection between the world we now live in, where the Fed pays interest on reserves, and the world we used to live in, where it did not. It goes without saying that Stephen does not bear responsibility for any perceived mistakes in the case I make.

I will argue that the Fed should raise the interest rate on reserves (IOR) and the federal funds rate (FFR) simultaneously, thereby keeping the opportunity cost of holding money at zero and enacting Milton Friedman's proscription for the optimal quantity of money.  In addition, the Fed should be given the authority to buy and sell an exchange traded fund (ETF) over a broad stock portfolio with the goal of achieving an unemployment target. This is an argument I have been making for some time but it is becoming more relevant as it becomes apparent that the world does not work in the way the New-Keynesians claim.

At the Jackson Hole conference this week, Janet Yellen argued for a broader range of Fed tools moving forwards. While I agree with that view, I would put a different emphasis on what those tools should be.  Maintaining a large stock of excess reserves is a good idea because money is costless for society to create. But, I argue, the Fed should actively trade the asset side of its balance sheet to promote real economic stability and a low unemployment rate. 


Two Interest Rates

Prior to 2008, commercial banks held reserves to meet a minimum required ratio of reserves to deposits. This ratio, the required reserve ratio, was mandated by the Fed and changed occasionally. In the period from the end of WWII up through 2008, banks held exactly this amount and excess reserves were equal to zero.

In October of 2008, the Fed began to pay interest on the  reserves of commercial banks held at the Fed. Figure 1, copied from the San Francisco Fed website, shows that, once that change in policy was enacted, the excess reserves of commercial banks increased from zero, before 2008, to approximately  $1.6 trillion in 2012. They are currently even higher at $2.2 trillion in 2016.

Prior to 2008, excess reserves were costly for commercial banks to hold because a commercial bank can make profits by lending reserves to private companies or to households or by purchasing interest bearing government securities. When three month T-bills pay positive interest, but reserves do not, a commercial bank will choose to hold zero excess reserves. After 2008, reserves held by commercial banks with the Fed are, to a first approximation, perfect substitutes for short term Treasury bills.

The post-2008 monetary environment is very different from the pre 2008 monetary environment. In the immortal words of Dorothy from the Wizard of Oz, “We’re not in Kansas anymore”.  In the brave new technicolor world, the Fed controls two money interest rates, not one; the interest rate on reserves held at the Fed (IOR), and the Federal Funds Rate (FFR). By raising the IOR and the FFR at the same time, the Fed can engineer an increase in nominal rates without altering the opportunity cost of holding base money. By keeping the difference between the IOR and the FFR at zero, the Fed is allowing the private sector to create what Milton Friedman called, the optimal quantity of money.

Because money is costless to create, from a social perspective, Friedman argued that society should set the opportunity cost of holding money equal to zero. That is precisely what happens when the IOR equals the FFR.  Because the demand and supply of money are identically equal, a simultaneous increase in the FFR and the IOR will not  lead to a contraction in the money supply.

The Real Effects of an Interest Rate Rise

Even if the opportunity cost of holding money is zero, a positive interest rate surprise may still reduce the money value of expenditures on goods and services as the values of existing dollar-denominated income flows are revalued downwards. This revaluation will not only affect fixed income dollar valued securities, it will also affect indexed bonds and corporate equities. The magnitude of that effect will depend on the public perceptions of future inflation.

In the past, we have observed a negative correlation between unanticipated increases in the money interest rate, and real GDP.  In 1979, the Volcker Fed engineered a big increase in the interest rate in order to bring inflation under control. It worked, and over the following decade, inflation fell.  But the interest rate increase triggered one of the largest post-war recessions in US history. The $64 trillion question is: can the Fed engineer a rate rise without triggering a recession? The answer to that question is yes.

New Keynesians believe that prices and wages were slow to adjust because of so called ‘menu costs’.  They incorporate that belief into their models with an equation that goes by the name of the ‘New-Keynesian Phillips Curve’. There is increasing evidence (Klenow and Malin) that menu costs are not large enough to explain the real effects of a monetary contraction and there is increasing evidence that there never was a Phillips curve in the sense in which the New-Keynesians need there to be one.

It is not prices that are sticky: it is expectations. The New-Keynesians are right to assert that the output gap and the unemployment rate are determined by ‘aggregate demand’. Aggregate demand depends on the beliefs of asset market participants about the future value of wealth and beliefs are self-fulfilling prophecies that depend on the ‘animal spirits’ of investors. We need a way for the Fed to intervene in the asset markets to stabilize asset prices and to prevent spillovers from volatile shifts in beliefs to the real economy.

One way to achieve that intervention would be to give the Fed the authority to trade in the stock market, as does the Bank of Japan. As I argued here, the Open Market Committee of the Fed should use that authority to actively trade an exchange traded fund in order to offset any potential drop in asset values that follows an interest rate increase. If unemployment is above target, the Fed would announce an increase in the growth path of the price of the ETF. If it were below target, they would announce a reduction in the growth path.

Why do I believe this would be effective? After all, conventional New-Keynesian wisdom holds that there is little or no connection between the value of the stock market and subsequent real economic activity. Conventional wisdom is wrong. Although there is little or no connections between short-run stock market fluctuations, I have shown in published empirical papers (2012,2015) that there is a strong and stable connection between persistent changes in the value of the stock market and changes in the unemployment rate. And I have shown in published theory papers, (2012,2013) why this chain is causal. When we feel wealthy, we are wealthy!

You can read more about these ideas in my book Prosperity for All, available for purchase on October 7th.