Economic News & Analysis—July 30, 2012

Fed options: Q&A

By Brian Jacobsen, Ph.D., CFA, CFP®, Chief Portfolio Strategist, and John Manley, CFA, Chief Equity Strategist

Will the FOMC engage in another round of quantitative easing?

I don’t think it should, but that doesn’t mean it won’t. I don’t believe that quantitative easing is likely to do the economy much good, though it could give the market false hope.

Why wouldn’t more easing do much good?

How monetary policy affects the markets differs from how it affects the economy. When Milton Friedman, a rather famous economist in the field of monetary policy, said, “Monetary policy operates with long and variable lags,” he meant that when the Federal Reserve (Fed) does something, it takes time—sometimes an unpredictable amount of time—for the policy move to affect the economy. I think that the amount of time it takes and the extent to which it’s effective depend crucially on what’s going on in the banking sector.

The Fed works with a small group of “primary dealers.” So if the Fed buys an asset, it does it through the Federal Reserve Bank of New York—and that bank deals with 21 other banks as counterparties. Basically, the Fed buys (or sells) from (or to) this group, and it starts the whole “money creation process.” That process looks a little gummed-up at the moment. It’s as though the Fed is providing “liquidity” to the banking system, but instead of being liquid like water, it’s liquid like maple syrup. I don’t think there’s any reason to expect that more liquidity would do more good.

What is wrong with the money creation process?

When the Fed buys an asset, it creates “reserves” for the bank that sold it. Those reserves may be lent out to customers or used to buy other assets. If the bank with the reserves lends them out, then that loan becomes someone else’s deposit at another bank. That deposit can be lent out, which becomes someone else’s deposit, and so on. In this way the “creation of bank reserves” becomes the “creation of money.”

At each stage of this process, banks have a minimum amount they have to keep “on reserve,” and that amount is the “required reserve.” Banks sometimes hold more than required, and this is called “excess reserves.” Currently, the banking system has a lot of excess reserves. Since December 2007, the amount has climbed alarmingly—by more than 84,000%. These excess reserves are called “excess” for a reason: Banks don’t need to hold them. As of October 2008, the Fed was given authority to pay interest on these reserves, as well as interest on required reserves. It’s not a lot—currently 0.25%—but it’s something. As long as the banks keep holding on to their excess reserves, the money creation process will remain sluggish.

Why doesn’t the Fed cut the interest it pays on reserves to get banks to lend?

That has been suggested. In fact, in October 2008, when it was announced that the Fed could pay interest on reserves, I wrote that if the Fed really wanted to kick-start lending, it should pay a negative interest rate on reserves. Basically, this would penalize banks for holding excess reserves. But now that so many reserves have been created, rate-cutting could be dangerous. I looked hypothetically at what would happen if banks lent out all that money, allowing it to go through the whole money creation process. I found that bank loans, which are just over $7 trillion as of the end of June, would skyrocket to $110.7 trillion, unleashing a massive increase in the price level. Because it takes time for reserves to be lent out and turned into deposits, which would then turn into other loans, and so on, the Fed would likely be able to slam on the brakes before loans reached a precipitous level. But it is a risk.

If you think about it, cutting interest rates on reserves would only deal with one side of the problem: loan supply. But the problem likely isn’t that banks don’t want to lend the money; it’s that creditworthy borrowers don’t want to borrow it. It takes two to make a loan: a lender and a borrower. The Fed wields a lot of power over lenders, but not so much over borrowers.

Quantitative easing seemed to work before. Why not try it now?

The first round of quantitative easing (QE1) was huge. It involved the FOMC expanding its balance sheet by buying up a lot of mortgage-backed securities. QE1 probably had a significant effect on pushing down yields on mortgages and Treasuries. QE1 started in 2008, and we may actually be seeing the fruits of that effort now, with the housing market beginning to turn around. It’s been nearly four years, and that may show how unpredictably long it can take for monetary policy to affect the economy.

The second round of quantitative easing (QE2) was about half the size of QE1 and involved buying Treasury securities instead of mortgage-backed securities. As a result of its size and its focus on Treasury securities, it probably influenced the economy and the markets less than QE1 did.

The FOMC then decided to give guidance that economic conditions would warrant keeping the federal funds rate exceptionally low for “an extended period of time,” which has since been revised to “late 2014.” This effectively outsourced some of the monetary easing to investors. Since investors had an expectation that rates would stay low until late 2014, they had little reason to invest at a rate that was any higher than that for loans to the government that matured before late 2014.

The FOMC then engaged in “Operation Twist,” which had the FOMC sell short-term Treasury securities and buy longer-term Treasury securities. This was supposed to “twist” the yield curve by pushing down longer-term Treasury yields. Recently, the FOMC announced that it would continue with Operation Twist until the end of 2012.

While it’s possible that the Fed could buy more assets, it might not find that desirable. First, in March 2009, the Fed entered into an agreement with the Treasury that the Fed shouldn’t buy assets other than Treasuries. Doing so was a form of fiscal policy, since it affected the allocation of credit instead of just the availability of credit. The Fed is supposed to handle monetary policy (that is, affect the availability of credit), and elected officials are supposed to handle fiscal policy (that is, the allocation of credit). Second, buying more Treasuries might not do much good. It seems to me that the yields that are most important to the economy have become somewhat untethered from Treasury yields. Third, the Fed has already expressed concerns that its asset purchases might be having undesirable side effects of the financial markets. Perhaps the Fed is trying to curb investors’ expectations that it will conduct more asset purchases. By highlighting the risks of more asset purchases, it might be sending an early signal to the markets that people should not expect it to do more.

Well, what can the Fed do?

I think there are three options left for the Fed. The first is to change the language it uses in its FOMC meeting statement. It could say that it will continue to target a long-term rate of inflation of 2% but keep rates low until the unemployment rate goes below 7%—as long as short-term rates of inflation stay below 4%. That could have a powerful effect on investors and the public’s expectations, while being consistent with the Fed’s long-term goals.

The second option that has been floated is to have the Fed provide loans through its discount window to banks, on the condition that the banks make loans. Instead of creating additional excess reserves, this would create reserves that would be lent out. This is actually an old policy of the Fed—based on an idea that dates back to 1705, with the writings of John Law—that has failed miserably. When the Fed was formed in 1913, it had something called the “real bills doctrine,” which would effectively allow it to lend money to banks, provided that the banks used the money to make loans to support “real” activity—that is, activity that supported manufacturing or trade rather than “speculative” activities. Originally, the Fed allowed unlimited lending, provided that the loans went to “productive” purposes. Sadly, this proved to be inflationary. The Reichsbank in Germany found that the practice could be hyperinflationary when it issued astronomical amounts of money to fund trade at ever-rising prices during the German hyperinflation of 1922 to 1923.

I’m not saying that allowing the Fed to lend to banks on the condition that the banks make loans would be hyperinflationary, but it might be useless. Remember, the banking system has a lot of excess reserves. Do banks really need more access to cheap money? Or is the problem lack of demand?

The third option is for the Fed to do nothing. This, in my opinion, may not be a bad idea.

Back to Recent Commentaries

  • Not FDIC Insured
  • No Bank Guarantee
  • May Lose Value