This archived content originally appeared at, the predecessor site to, and does not carry the sponsorship of the Cato Institute.

Fed needs to stop asset acquisitions for a generation or so

The Federal Open Market Committee (FOMC) meeting that ended today (Oct. 29) marked the first chance for the FOMC finally to do the right thing since the onset of the great financial crisis in the late summer of 2008. That right thing consists of resolving not to add even another dollar to the Federal Reserve System's balance sheet for at least the next ten years (and perhaps as long as 30 years) in the absence of officially declared war or national emergency. Thankfully, on an 11-1 vote, the FOMC finally adopted the initial step in that policy direction, agreeing not to make significant additions to the System's securities portfolio, for the time being.

The great financial historian Charles P. Kindleberger (1910-2003), who taught at Massachusetts Institute of Technology throughout the postwar years, was struck by what he perceived as the tension between generally Keynesian monetary policy (ignoring quantities of money and focusing instead on interest rates and unemployment rates) and generally monetarist monetary policy (giving great importance to measurement of quantities of money, tax policy, and sustainable economic growth, with the market sorting out interest rates and unemployment rates). In his Keynesianism vs. Monetarism and Other Essays in Financial History (1985), Kindleberger wrote, essentially, that long periods can pass when Keynesian policies may be pursued with benefit or at least without noticeable harm but that, when the cycles turn and the monetarist policy becomes appropriate, the monetarist approach is "so very timely." Here, reference to monetarist approaches should be understood to be attention to the quantity theory of money: Many Austrian-school economists and even some traditional Keynesians care about and pay attention to the quantity of money.

Thus, the FOMC majority could have concluded today that a Keynesian approach to the financial crisis had a nice, nearly seven-year run but that, with clear statistical evidence of diminishing benefit from the Fed's experiment in expanding reserves to levels well in excess of anything that Kindleberger would have considered wise, it is time to stop. From here on out, probably for ten years or longer (perhaps up to 30 years), the FOMC should pursue monetarist approaches to policy in which, for every dollar of assets added to the System's portfolio, another dollar is sold from that portfolio, even during emergency periods, and in which maturing assets are not replaced, with net shrinkage of the portfolio over time. The FOMC did not adopt this last policy step today, voting essentially to hold the size of the portfolio constant until further notice.

One cannot argue plausibly that necessary market liquidity would be reduced below sustainable levels by attention to the quantity of monetary base that the Fed creates. (Domestic monetary base = currency in circulation plus reserve balances held at the Fed; foreign exchange swap drawings in dollars, currently zero or near-zero, should be added to this amount to find total probable domestic claims against the Fed.) Currently, there are about $1.25 trillion of currency outstanding (with probably about 70 percent held outside our borders), plus about $2.7 trillion of reserve balances held at Reserve Banks. That is nearly $4 trillion of monetary base.

In 2007, the year before the crisis, a Fed balance sheet of "only" $929 billion sufficed to promote strong growth in a $14.5 trillion economy (nominal GDP). The Fed's balance sheet was only 6.3 percent of the entire economy. After countless interventions in the economy and a never-ending series of Quantitative Easings (econospeak for money-printing) since then, the Fed's balance sheet is nearly five times larger, but the economy is only 19.3 percent larger. The Fed's balance sheet is now 25.5 percent of GDP.

One supposes that it takes a lot more money to make the world go around these days, but the economic outcome is far smaller than one would have expected given the amount of monetary input. If the Fed has an econometric model showing how much GDP growth it expects from each new dollar of monetary input, it should disclose that model to Congress now, and if the outcomes are suboptimal or as demonstratively inefficient as I think they are, then Congress should make the Fed stop using that model to drive FOMC policy choices, if the Fed refuses to do so voluntarily.

The Fed courts a real danger of becoming, if it has not already become, the motor of a thoroughly corporatist political economy model for the United States, if not for the entire world. A central bank balance sheet equal to 25 to 50 percent of GDP was considered a hallmark of corporatism in developing economies that the World Bank was trying to reform in the post-1980 years. The Fed should be asked to tell Congress now, before the election next week, how great a percentage of GDP it wishes to hold on its balance sheet without seeking the approval of Congress.

Back to Kindleberger's point: When the time comes around for the monetarist message, it is important for central bankers to heed that message. It is, indeed, time to stop printing money (technically, this is a collaborative exercise involving both the Treasury and the Fed and, behind the scenes, both the White House and Congress).

The following facts are clear: As of mid-2014, the Fed had expanded its balance sheet by $3.483 trillion since August 2007 (375 percent), with nearly all of the increase occurring since the onset of the crisis in September 2008. However, nominal GDP expanded by only $2.850 trillion over the same period (19.3 percent). In other words, only 81.8 cents of new GDP were created for every dollar of Fed-Treasury money printing, an exercise of remarkable inefficiency considering that, for the eleven years before the crisis, 1997-2007, about $13.88 of new GDP were created for every new dollar of money printing. Money printing is an inefficient way of creating GDP, after the crisis, but it has proved to be an efficient way of creating asset price bubbles.

Finally, if one wished to reduce the Fed's role in the economy to the level that prevailed before the crisis, about 6.5 percent of GDP (the range was 5.9 to 6.9 percent over the preceding eleven years), the current size of the Fed's balance sheet would support economic expansion to nominal GDP of $67.9 trillion, about four times the current size of GDP. Historically, it took 15 years for GDP to quadruple, 1969-1984, and that period included the high-inflation 1970s. In a period of lower inflation, after 1984, it took 28 years for GDP to quadruple again in 2012. That is why I proposed, at the beginning of this note, that we simply suspend the monetary policy operations of the Fed for a generation or so until the rest of the economy catches up to all the monetary base that recent Fed operations have created.

We still need banking supervision for as long as we have non-gold fractional reserves, we need the payments mechanisms operated by the Fed, and someone has to buy all that debt that the Treasury has for sale. But it is not clear that the Fed is the entity that should do any or all of these things. On the other hand, we have a large infrastructure investment in the Fed, and we might choose to keep it operating to perform these other functions. Just not monetary policy, not for a good long while, anyway.