Bernanke turns on the printing press: John Kemp

-- John Kemp is a Reuters columnist. The views expressed are his own --

LONDON (Reuters) - By purchasing long-term U.S. Treasury securities, the Fed has confirmed that its solution to the debt crisis is more of the cheap money policies which caused the problem in the first place, and risks reigniting commodity prices.

It is a curious irony, but some prices are simply too important to be left to the market. In the advanced economies, where the price of most goods and services is determined by supply and demand, the price of money itself (i.e. interest rates) is still fixed by the central bank.

Since the late 19th century central banks have routinely manipulated the price of money, buying and selling government securities and commercial debt in the open market to influence their price and yield, a process known as open market operations (OMO).

The 1913 Federal Reserve Act gives the Fed a very broad remit, allowing it to undertake open market operations in long-term as well as short-term securities, and in debt issued by private institutions as well as the government.

Since 1953, the Fed has largely restricted itself to conducting operations in government rather than private debt, and mostly in securities at the short end of the curve. The Fed has sought to control near-term rates, leaving the determination of long rates to the market.

In response to the crisis, the Fed has already reversed the focus of open market operations by starting to buy private debt from the market in what are in effect open market operations. Now the Federal Open Market Committee has announced it will buy $300 billion of “longer-term” Treasury securities over the next six months, moving open market operations further along the yield curve.

By influencing the price of securities, open market operations affect the price of money (interest rates). Because the Fed essentially buys securities by crediting sellers with claims on itself, OMO can also create additional money (the modern electronic equivalent of the printing press).

In general, the Fed has been careful to limit increases in the money supply from open market operations. Its portfolio of Treasury securities increased slowly from $490 billion in June 1997 to $591 billion in June 2002 and $790 billion in June 2007. Most of these are short-term bills maturing within twelve months. The aim has been to grow the money supply at slow but steady rate and provide a gradual increase in liquidity in line with the rise in economic activity.

The Fed’s decision to purchase $300 billion of “longer-term” Treasury securities is more radical. The Fed will be buying just 5 percent of the outstanding stock of U.S. government debt from the market (far less if debt the federal government will need to issue later this year and next to cover the cost of rescue operations and its deficits is included).

But the government's debt profile is heavily weighted toward the short end. Of $5.975 trillion of U.S. government debt in issue to the public and institutions at the end of February 2009, $3.965 trillion (66 percent) was short-term debt due to mature before the end of 2012. Another $966 billion matures between 2013 and the end of 2016. Just $1.043 trillion matures in 2017 and beyond (here).

The Fed has not yet specified what it means by “longer-term”. But assuming it focuses on buying debt with a maturity of ten years or more, then $300 billion will amount to 28-34 percent of the total outstanding stock of debt at these maturities.


By concentrating large-scale purchases in a relatively thin and illiquid part of the market, the Fed will have a huge impact on the price of these securities and benchmark interest rates, which is precisely what officials intend. But the policy is fraught with risk (one reason the Fed has hesitated until now):

(1) The Fed risks creating a massive bubble in long-term Treasury debt, becoming the buyer of last and only resort for securities the rest of the market considers over-priced.

The federal government still needs to issue trillions of dollars of new securities to fund its forecast deficits in 2009 and 2010. Unless it can issue more debt at the back end of the curve, the constant need to refund expiring short-term securities will make the government prisoner of shifting market sentiment. It will also complicate the conduct of monetary policy.

So the government needs to start issuing more long-term debt. But if the Fed drives up long-term debt prices to an artificially high level, the government may struggle to find private buyers, and the Fed could end up having to absorb most or all of the new issues. In effect, the Fed would be monetizing the government’s deficit.

(2) The Fed risks creating an incentive for foreign central banks to liquidate their holdings of long-term U.S. securities and convert them into other instruments in other currencies. Most Treasuries at the back end of the curve are held by pension funds (liability matching) and foreign central banks (investing their foreign exchange reserves). China and the other central banks are the largest holders of Treasury debt. While the exact maturity profile of their holdings is unknown, it will include large numbers of long-term notes and bonds.

So far, China’s State Administration of Foreign Exchange (SAFE) has been trapped in the Treasury market. The Fed has, however, provided a (partial) exit strategy. It has created a guaranteed buyer in the market with non-commercial motives prepared to pay an artificially high price for the securities. China can sell some of its holdings directly or indirectly to the Fed, while the Fed supports prices. In effect, the Fed will end up buying some of the Treasuries China no longer wants.

In the end, the Fed’s long-term open market operations will probably prove futile. The impact of the $300 billion purchase program (which will drive down back end yields) risks being swamped by the larger volume of new debt that the federal government will have to issue at these maturities (driving yields up).

To keep rates down, the Fed would need to keep expanding the program almost indefinitely.


Long-term open market operations will also have effects on the rest of the economy (some intended, some not):

(a) The unexpected aggressiveness of the Fed’s move, after it seemed to back off from the proposal in recent weeks, coupled with its willingness to monetize part of the government’s long-term debt, sends a powerful signal to investors. Stimulating recovery will be the Fed’s over-riding, and only, priority for the next 2-3 years. Officials are willing to risk a medium-term rise in inflation, including a rise in commodity prices, as the price of restarting growth.

The Fed has created a strong incentive for investors to increase exposure to physical commodities and commodity futures as a way to profit from the reflation play. Prices for gold, crude oil and copper have all risen sharply in the 24 hours since the Fed announcement, as the market anticipates a more reflationary environment over the next 2-3 years.

(b) By pushing already-low yields on “safe” assets such as government bonds even lower, the Fed is trying to force investors out of these safe havens and into riskier asset classes if they want positive returns. Adjusting for current and likely future inflation, the Fed is pushing real yields on government debt to zero or below it. In an environment characterized by (moderate) inflation, investors will be forced into other, riskier asset classes if they want to protect the value of their capital.

The Fed is repeating the cheap money strategy which it used in 2001-2004, using ultra-low interest rates to push investors along the risk curve from government bonds to corporate bonds, equities, real estate and commodities in search of yield. But by artificially depressing yields on government bonds, the Fed risks creating the same asset mispricing as before.

It risks creating an even larger bubble in the future, this time centered on the government bond market. But that seems to be a risk officials are willing to take at present, and worry about correcting later.