Thursday, September 23, 2010

Treasury TIPS -- Treasury's Motivations

Perhaps the most contentious issue I had to deal with during my tenure at Treasury was the controversy over issuance of inflation-indexed bonds. It certainly was the longest lasting. Now a new paper ("Why Does the Treasury Issue TIPS? The TIPS-Treasury Bond Puzzle") by three professors at the UCLA Anderson School written for the National Bureau of Economic Research (Mattthias Fleckenstein, Francis A. Longstaff, and Hanno Lustig) argues that TIPS (Treasury Inflation-Protected Securities) are a costly form of finance and asks why the Treasury "leaves billions of dollars on the table by issuing securities that are not as highly valued by the market as nominal Treasury bonds." (A September 6 draft version of the paper is available for download here.)

When it comes to academics, Treasury just can't win on this issue. It was, after all, academic economists who were the loudest proponents of Treasury issuing the bonds. Proponents included such luminaries as James Tobin, Milton Friedman, and Stanley Fischer. Now that Treasury has been issuing TIPS for almost 14 years, Treasury is being attacked by academics for issuing these securities.

During the Reagan Administration, there was a big push by some Administration economists to get the Treasury to issue inflation-indexed securities. The Domestic Finance section of Treasury was consistently opposed. One senior political appointee joked that the way to market inflation-indexed bonds was to use the membership mailing list of the American Economic Association. Others at Treasury, though, were strong proponents, including Under Secretary for Monetary Affairs Beryl Sprinkel, to whom the Assistant Secretary for Domestic Finance reported (the organizational chart at Treasury has since changed.). However, even with that high level support for inflation-indexed bonds, the proponents were never able to convince the various Treasury Secretaries to issue this new type of security.  It is useful to remember that at the time of these debates, inflation had been high but was being brought under control.

In the George H.W. Bush Administration, the issue receded at Treasury even if academics wanted to pursue it. The political leadership of Domestic Finance at Treasury was just not interested. (Interestingly, Vice President Dan Quayle had been a proponent of inflation-indexed bonds as a Senator and raised the issue at a hearing of the Joint Economic Committee, but, as far as I know, he did not press the issue as Vice President.)

During the Clinton Administration, inflation-indexed bonds became a live issue again. Two strong proponents of inflation-indexed bonds were Larry Summers and Alicia Munnell, both of whom had an academic background. Fed Chairman Alan Greenspan was also a strong proponent. Secretary Rubin, reflecting the consensus view of the major government security dealers at the time, was initially dubious, but eventually he became convinced. The political leadership of Domestic Finance was more ambivalent about the issue than their predecessors had been, and, in any case, taking on Larry Summers is not something one does lightly. When it became clear that Rubin was likely to decide to issue inflation-indexed bonds, I began work on the technical details.

The case against inflation-indexed bonds was mainly based on doubts that they would be cost-effective from Treasury's point of view. Reason for these doubts included: (1) lesser liquidity than conventional Treasuries; (2) limited demand because the appreciation of principal would be taxed currently even though it was not paid out; and (3) no evidence that inflation indexation was something for which there was a lot of demand (efforts to issue price level adjusted mortgages had not been successful).

The proponents' main argument was that inflation-indexed bonds would be cost-effective because investors would be willing to pay up for the inflation insurance these securities offered. In addition, proponents said that the bonds would act as a "sleeping policeman," because Treasury's interest costs would soar on these securities if inflation got out of control. The bonds also could motivate the private market to come up with more inflation-indexed products, such as inflation adjusted annuities (though some, but not all, opponents viewed this as a drawback because it would reduce the size of the anti-inflation constituency.)  In addition, the bonds would provide both the market and policymakers information about inflation expectations which would lead to better decisions.

During the initial years of TIPS issuance, it was clear that the securities had not been cost-effective up to that point. There was always hope that as the market grew and became more liquid, the pricing Treasury received at its auctions would improve.

At this point, though, getting rid of TIPS would be difficult for Treasury to do. A surprise announcement, such as Peter Fisher's announcement of stopping the issuance of 30-year bonds, would not be good policy. Moreover, the Treasury has made numerous statements confirming its commitment to TIPS, and any sudden reversal of policy would be viewed as breaking faith with the market. For example, the sponsors and investors in TIPS mutual funds would be very unhappy if these securities were no longer issued.

While a gradual reduction of TIPS issuance is a possible policy course, I think that Treasury career staff would be hesitant to recommend major changes to the TIPS program on their own initiative, given what a hot potato this subject has been. In any case, I do not know if they agree with the UCLA authors that the program is currently an expensive form of financing. Any decision to review possible major changes in TIPS issuance would have to come from the political appointees. Even if they were inclined to reduce substantially or eliminate TIPS issuance, I think most political appointees would be hesitant to face the criticism that would be hurled at them on this particular issue.

If the authors of the NBER article are correct that TIPS continue to be expensive, there are reasons why Treasury is likely to continue with the program. What the authors do not discuss but is an interesting question is how TIPS will fare if the U.S. enters a long period where inflation is negligible. At the moment, even though there are deflation fears, there is also a concern about he eventual reemergence of inflation. No one knows if the fear of future inflation will lessen, but, if it does, that would certainly impact the TIPS market.

Monday, September 20, 2010

The Maturity Structure of the Public Debt

Over the decades there has been a debate about the maturity structure of U.S. Treasury debt. At one time, not only was there a debt limit but also a limit on the amount of bonds Treasury could issue with an interest rate greater than 4 percent. Bonds are similar to notes but with longer maturities. Currently, the dividing line is 10 years; it has in the past been shorter. When Treasury was given room in the late 70s and 80s to issue more bonds without regard to the 4% ceiling, Treasury officials took great pride in growing the market for 30-year bonds and were distressed during the Reagan Administration when some political appointees wanted to halt this progress by shortening the maturity of the public debt.

This did not happen during the Reagan Administration, but in the latter part of the Clinton Administration the issuance of 30-year bonds was reduced, and then it was eliminated for a time during the George W. Bush Administration.

Now Treasury has reversed course. In a letter to the Wall Street Journal published on September 18, Assistant Secretary Mary J. Miller proudly states: "We have explicitly pursued a strategy of reducing reliance on short-term debt. Over this period, the average maturity of the debt has extended at the fastest pace in history—from 49 months to more than 58 months today."

There is no easy answer as to the proper maturity of the public debt, and, after hearing and sometimes participating in arguments about this over many years, I have come to the conclusion that it may not matter that much as long as one does not follow extreme policies. I think, thought, that it is good policy to issue 30-year bonds, since different maturity sectors attract different investors, and Treasury needs as many investors as it can get. Also, it is useful for the economy to have a long-term benchmark interest rate.

I should note that the arguments about the maturity structure often focused on the average maturity of the public debt, which masked how sensitive interest costs are to short-term interest rates. Miller, in her letter, states that 55 percent of the public debt matures with three years. It is not clear whether she is referring to the total public debt, that held outside of government accounts, or that held outside of government accounts and the Federal Reserve system. But no matter. A change in interest rates will impact the budget fairly quickly.

In the Bush Administration, two arguments were used to justify the elimination of the 30-year bond. The first argument, at the beginning, was that the budget was in surplus and we did not need to issue long-term debt. There was, unbelievable as it may sound now, a concern at the end of the Clinton Administration and the beginning of the Bush Administration that the surpluses would continue for such a long-time that the Fed would have to look to other securities with which to conduct open market operations. (I remember thinking at the time how misplaced this concern was; budget surpluses would not continue because they were politically unsustainable and a recession, which at some point would happen, would cause it to go away. The Congress would either spend it or enact tax cuts, probably both. This in fact did happen, along with expenditures for wars in Iraq and Afghanistan that I did not foresee.)

It soon became clear in the Bush Administration that the budget surplus rationale for eliminating the 30-year bond was not valid. The argument then shifted to the shape of the yield curve; since it is usually positively sloped (that is, long-term yields are higher than short-term yields), the Treasury could over time save money by issuing only short-term debt. Note that this argument pushed to the extreme would mean that Treasury should only issue bills, though no one advocated that.

In the event, it was Under Secretary Peter Fisher who made the decision to eliminate the 30-year bond and it was reversed after he left, though not immediately. As I recall, the Treasury made some convoluted and unconvincing arguments that this was not a change in policy, but of course it was.

A flaw in the argument that Treasury could save money over the long-term by issuing mainly short-term debt was revealed by this episode. Political appointees are at Treasury for the short-term, and cannot stop their decisions from being reversed. (Of course, some decisions are easier to reverse than others.)

The current repudiation of the shortening strategy begun in the Clinton Administration and pursued with a vengeance at the beginning of the Bush Administration is striking. Treasury is probably right to pursue the current policy.

I am of the school that says that Treasury debt management should normally be very boring except to traders. The Treasury is too big to try to beat the market consistently. As one Assistant Secretary in the George H.W. Bush Administration remarked, it 's tough for an elephant to dance.

Simon Johnson and Peter Boone -- "Brady Bonds for the Eurozone"

I was surprised to see that Simon Johnson and Peter Boone are advocating "Brady bonds" for countries such as Ireland and Greece.  Their article on this can be found here, and a useful March 2000 "primer" on Brady Bonds (published by Salomon Smith Barney) is available here.

The reason I am surprised is that these bonds are based on an accounting trick designed to mask what was really going on.  I hope Simon Johnson and Peter Boone are not advocating this because of the accounting.

Brady bonds were named after Treasury Secretary Nicholas Brady, who advocated an exchange of bank loans to developing countries in financial trouble for these bonds.  This program began in the late 1980s and continued into the 90s.  

The  principal of a Brady bond was collateralized by a zero-coupon Treasury bond.  In many cases, the Treasury issued this zero-coupon bond as a non-marketable special issue; in other cases, the zero-coupon collateral was obtained in the market by buying Treasury STRIPS.  (There was an internal dispute early on about the pricing of the non-marketable zero-coupon bond issued to Mexico, which became public and was both the subject of a Congressional hearing and a GAO report.)

Because the principal was collateralized by a Treasury security, the banks could hold these bonds at par on their balance sheets.  However, the value of a long-term bond is more dependent on the periodic interest payments than it is on the principal payment.  This was the accounting trick.  The loans had to be marked down but the bonds would not have to be, even if the payment of interest was subject to significant credit risk.  For the country restructuring its debt, a 20-year or 30-year zero-coupon Treasury could be obtained quite cheaply at the interest rates prevailing at the time.

I am not sure whether this scheme would work now, since accounting standards have evolved to force more mark-to-market valuations, and zero-coupons are of course more expensive for a given maturity currently because of the very low interest rate environment.  I also am skeptical of a policy relying on accounting mirage.

Simon and Boone need to make a better case addressing these points if they want to be convincing in their advocacy of Brady bonds.

Thursday, September 9, 2010

Interest Rates for Short-Term Savings -- Money Market Mutual Funds and Bank Accounts

The current low rates of interest are painful for those with cash to invest, and particularly painful for those living off their investments in retirement.

The interest rates obtainable in money market mutual funds is currently pathetic. For example, the year-to-date rate as of September 9 of Vanguard's Prime Money Market fund is 0.04%. Though there has been an outflow, inertia has served to keep many individual invested in the money market funds. Banks, though, are offering much higher, though historically low, rates on various types of savings accounts and CDs. A quick Google search will provide links to online banks offering around 1.3% interest for savings accounts with a transaction limit of six withdrawals per month. Also, the bank accounts are safer than the money market mutual funds, because they benefit from FDIC insurance. The federal government's insurance for money market mutual funds, which was provided for existing accounts during the 2008 crisis was quietly ended.

This rate disparity is reminiscent of the late 70s and 80s, when "disintermediation" was the word bankers hated to hear. The popularity of  money market mutual funds began in the 1970s -- a period during which banks and savings and loans were restricted in the rate of interest they could pay by Regulation Q. These funds essentially helped save the mutual fund industry during a period when individuals were not interested in investing in the stock market. Money flowed out of banks and into money market mutual funds in a period of high and increasing interest rates.

Now, the reverse seems to be happening in this new, low interest rate environment. Banks are providing higher rates than the competition, and, if this persists, money market funds will shrink much more than they already have. Individuals will find the bank CDs and savings accounts attractive, especially since this is now a period where one does not fear looking stupid in avoiding the stock market. For individuals, the bank accounts also look good since these government-backed accounts provide higher rates than Treasury bills. One-year T-bill rates are currently 0.24%, and shorter-term bills have lower rates.

Of course how long banks will be able to offer higher rates depends on their ability to invest this money, such as in new loans, at a profit. Part of what is going on is that some banks are trying to obtain funding that will be cheap over the long term, if not currently, since demand deposits and savings accounts are sticky once established. But if the flow into bank accounts pick up to an extent that the banks cannot profitably use the money, the relative attractiveness of the rates they offer will have to diminish.  

There does not seem to be much written about the macroeconomic effects of this.  I suspect that analysts looking at this are not sure what to make of it, just as economists, in general, are at sea about the likely course of the economy.

Deflation or inflation?  Recession, depression, or slow recovery?  The elaborate economic models of some forecasters are not of much guidance since no one has experienced the current mix of economic factors. Government policymakers cannot rely on models, and  policy will be the result of a combination of judgement and politics. Currently, politics appears to be clouding judgement.