Comments on Current Conundra

April 26, 2007 Austin, Texas

Thank you, Luther [King]. You are a generous man. Luther chairs the ad hoc Financial Advisory Group that I occasionally convene at the Dallas Fed and is an invaluable advisor. But his and Teresa’s friendship with the Fisher family goes way back: Teresa was the deaconess who guided our daughter Texana through her confirmation at Highland Park Presbyterian Church a long time ago—in fact, in the last century. Our boys know each other and are friends. And from a professional point of view, I have admired—and been jealous of—Luther since I first started up Brown Brothers’ operations in Texas almost 30 years ago. He is a man of punctilious courtesy and the nicest sense of personal honor and about as decent and forthright a friend as anybody could hope to have. I am honored to be here with you today, Luther.

I used to be one of you. As Luther kindly mentioned, I ran a little investment advisory firm and hedge fund way back in the days when the clients and limited partners made more money than the investment advisors and general partners did. Imagine that!

I had planned to speak today about four conundra: the impact of globalization; the problem with relying on the data that you and I and we all look at to gauge economic performance; how you—and I use the term “you” deliberately, as you will see later in this talk—will deal with the problems of fiscal recklessness that have led us into a cul-de-sac of long-term liabilities of almost unfathomable dimension; and what a former distinguished associate used to refer to as the yield-curve-shape “conundrum.”

Upon reflection, however, I realized that the latter is really not much of a conundrum. Let me explain, employing the caveat that I am speaking today, as I always do, solely in my personal capacity and am in no way speaking on behalf of the Federal Open Market Committee or for anybody else in the Federal Reserve.

There is an enormous amount of liquidity coursing though the arteries and veins and capillaries of the financial system worldwide. There is only so much of that liquidity that can be placed by responsible fiduciaries in lesser credits since the markets for non-dollar, non-euro, non-pound denominated paper are limited. This is not to say that investors have declined to invest in other markets, as we see clearly in the narrow spreads between the biggest and most established credits and lesser ones.

Besides, the most prominent central banks have been exemplars of good behavior: the rectitude of the Fed, the European Central Bank and the Bank of England is, I think, above question. (As, incidentally, has been the behavior of the Aussies, the Canadians, the Mexicans and many others). The Fed and other central banks have gained credibility in keeping inflation low and stable.

Better inventory policy has helped make real rates more predictable by smoothing out business cycles. Pension funds are—surprise!—actually matching their investments to the maturity schedule of their liabilities. And in a rapidly globalized world, foreign demand for Treasuries has increased both for portfolio diversification needs and the desire of some newly flush central banks to have ample reserves to reduce currency volatility.

Declines in risk premia are therefore understandable against the background of sustained, robust growth worldwide, the concomitant reduction in volatility of the global macroeconomy and less inflation risk. And the use of derivatives has reduced the short-term price risk of long instruments, including Treasuries. So, where is the conundrum? Why is it so puzzling that longer rates are relatively low?

Now, to be sure, one might try to dispense with the “riddle” of the yield curve by arguing that the reason for having a flat to negatively sloped yield curve is that the market is forecasting the possible onset of recession.

It is true that inverted yield curves have historically presaged recession, but with one rather interesting exception that old folks like Luther—let me restate that—that the historians among you—will recall. In 1967, we had a credit crunch hit the housing sector particularly hard and we were fighting an unpopular war in a far-off land. The yield curve flattened, then inverted, yet no recession ensued.

That said, I think that as we progress deeper and deeper into a globalized marketplace with more sophisticated methods for hedging risk, and with central banks working double time, overtime, to exorcise the demon of inflation, there are reasons other than concern for the future of the economy for investors to lower the risk premia they have historically demanded as they move out along the yield curve.

Which leaves us with the other three conundra.

By now, many of you know that globalization is a preoccupation of the Dallas Fed. We are building the Globalization and Monetary Policy Institute. We orient most of our research and prepare for Federal Open Market Committee meetings by looking at the world through a global lens.

We surmise that the integration of markets for goods, services and capital has been facilitated by the physical and cyber linkage of the planet. That integration has changed the gearing of the U.S. economy and brought into question many, if not most, of our most treasured economic and monetary conventions.

We see tremendous behavioral shifts in the real world as middle managers everywhere take advantage of increased global integration by reaching across physical borders and though cyberspace to drive down their costs of goods sold and their G&A costs, tighten their inventory management, improve supply-chain efficiency, enhance productivity and access new markets.

We note that this goes deeper than the simple trade in goods. And we know that it affects not-for-profit enterprise as readily as for-profit enterprise.

Let me give you a simple example. One of the great lung specialists in the world is Dr. Jonathan Weissler, the chief of medicine at UT Southwestern in Dallas. After he sees a patient, he dictates his notes into a wireless voice recorder. The recording is transmitted electronically to a service that employs English-speaking scribes all over the world, often in India. When Dr. Weissler comes to work the next morning, there on his desktop is a transcript. So while the good doctor gets his 40-winks-sleep overnight, someone in India has written up his notes at a fraction of the cost of having them transcribed locally and—of concern to those worried about the status of education here in the U.S.A.—with greater accuracy. Dr. Weissler’s productivity is enhanced. With the savings that come from utilizing globalization, he can put more time and money into saving lives here at home.

We see this pattern repeated over and over again, countless times in every size and shape of enterprise. So what is the conundrum here? What riddle does globalization’s impact pose? Well, the riddle, or more appropriately, the question, is whether or not this impacts how the Federal Reserve executes policy in accordance with our dual mandate to foster the monetary conditions necessary to foster noninflationary sustainable employment growth.

How do we deliver on this mandate in a globalized world? How do we monitor global capacity constraints when statisticians in other countries measure things differently and at different time intervals? What instruments should we use to determine the optimal speed that our economy can grow in a rapidly integrating world? In economists’ terms, what is our NAIRU—our non-accelerating inflation rate of unemployment? What is its impact on inflationary expectations? How is globalization conditioning the behavior of workers? Of consumers? Of capital?

No one really has the answers, although some pretend to. So we are working hard at the Dallas Fed and throughout the Federal Reserve to come up with new data and new models that fit the new world.

Which brings me to the next conundrum: the phenomenon of “data dependency.” Some time ago, I gave a speech titled “Confessions of a Data Dependent.” This was back when it was in vogue to say that monetary policy was data-dependent. And yet, upon reflection, I have personally come to feel that this is a bit of an oxymoron, a contradiction in terms, like “jumbo shrimp” or “instant analysis.” Why? Because so much of the data you and I consider essential are moving targets, subject to constant revision—perhaps because the new, globalized gearing of the economy makes measurement so difficult. How “data dependent” can we really be?

In the 1920s, there was a British Inland Revenue agent who later became a director of the Bank of England named Josiah Charles Stamp. Stamp once quoted a friend who had observed, “The Government [is] extremely fond of amassing great quantities of statistics. These are raised to the nth degree, the cube roots are extracted, and the results arranged into elaborate and impressive displays. What must be kept in mind, however, is that in every case, the figures are first put down by a village watchman, and he puts down anything he damn well pleases!”

America’s statisticians are a careful lot who do not record anything they darn well please. But it is important to remember that even our best measures of economic performance are subject to substantial revision well after the fact, not only because the modern equivalent of the village watchman occasionally puts down faulty numbers, but also because we can’t always get numbers from all the watchmen in real time.

There is a tension between timely data and accurate data. Timely data are often based on estimates and probabilities that are injected to fill in missing numbers, only to be changed when more complete information comes in. Markets react to the timely data with apoplectic frenzy, only to have the same number quietly revised many months or even years later in light of more substantiated evidence, but without the same market fanfare.

Let’s go to the videotape on inflation. In the 1990s, concern grew that our main inflation gauge, the Consumer Price Index, was providing a distorted view of price trends. A pickup in the pace of productivity growth in some sectors and an expansion of the reach of global markets due to the far-reaching trade-liberalization policies of Presidents George H. W. Bush (“41”) and Bill Clinton were, together, leading to sharp declines in the relative price of goods like apparel and consumer electronics. The CPI reflects the resultant shift in household spending patterns with a substantial lag and, in the meantime, puts too little weight on falling prices and too much weight on rising prices. The more quickly household spending patterns change, the greater the likelihood of upward bias in CPI inflation.

An alternative inflation gauge—the deflator for personal consumption expenditures—both continuously adjusts for changes in the composition of household spending and also has broader coverage than the CPI. So, in 2000, federal policymakers adopted the PCE price index excluding food and energy as their preferred measure of inflation trends.

Unfortunately, the nice theoretical properties of the PCE inflation measure come at a price: PCE inflation is not released until several weeks after CPI inflation (which confuses the public) and is revised, often substantially, when new, more complete estimates of the composition of household spending become available (which confuses the analytical community).

Core inflation for 2003 was initially reported out at 0.9 percent. Later, having gone back and studied the entrails, the Commerce Department put inflation during 2003 a full half percentage point higher, at 1.4 percent. Similarly, 2004 core inflation was initially thought to be 1.6 percent, but subsequent revisions put it at 2.2 percent.

Or look at GDP estimates. The first release received for GDP growth for the fourth quarter of last year was 3.5 percent. A month later, it was revised downward to 2.2 percent. Recently it was revised upward to 2.5 percent.

Remember this when the numbers for the first quarter of this year are released tomorrow. GDP growth, industrial production, retail sales and payroll employment are imperfect statistics that make a splash when they are first released and yet hardly make a ripple when their true nature is finally revealed.

Data revisions have important implications for policymaking and policy evaluation. Potentially, they can erode central bank credibility. Partly because of data revisions, it is hard for us to spot turning points in the economy. The fact is, statistical agencies fill in missing data with extrapolations that are especially likely to be wrong at these turning points because the estimates and probabilities they use are biased toward the latest trend. The result is that we are likely to underestimate slowdowns and pickups at precisely the moment when we need to take corrective action.

Inevitably, a monetary policymaker’s reliance on data that are subject to revision means that some decisions would have been marginally different with the benefit of hindsight. I say “marginally different” for two reasons. First, the FOMC is eclectic in that it monitors a broadly diversified portfolio of indicators and reports from a variety of sources in the expectation that the noise elements in the various indicators and reports will tend to cancel one another out. This portfolio approach limits the weight on any one piece of information or any one source. Second, our policy decisions are influenced by what’s happened in the past mostly through how the past conditions the outlook for the future: We’re focused on where inflation and real growth seem to be heading, more so than on where they’ve been.

In this respect, Fed policymakers are value investors rather than momentum traders. After hard experience, we’ve learned how to deal with revisions and are able to manage them more or less successfully. We don’t radically alter policy based on inflation or output realizations as long as the reasoning behind our original assessment of the economic outlook seems sound. In a sense, we have learned the dangers of taking numbers based on estimates and probabilities at face value alone.

Estimates and probabilities bring me to the last conundrum: How you are going to save your children and grandchildren from facing the certain probability that the massive liabilities accumulating to our Medicare and Social Security programs are going to rob them of their futures.

According to the latest official U.S. government trustee reports that were released on Monday of this week, the infinite-horizon discounted present value of our unfunded liability from Social Security and Medicare—in common language, the gap between what we will take in and what we have promised to pay—now stands at $88.2 trillion. The potent combination of lower birthrates, higher medical costs and longer life expectancies provides little reason to hope the figure will fall. Last week, I shared my concerns about our long-term liabilities that were based on earlier trustee reports, which tallied the shortfall at $83.9 trillion, a full $4.3 trillion less that this new report suggests.

Just how big is an $88.2 trillion shortfall? Well, it is almost seven times the U.S. gross domestic product. It is more than 100 times the country’s annual defense budget. If you divide the $88.2 trillion evenly among the 302 million U.S. residents, you get a per-person liability of $292,000—more than six times the average household’s annual income. Each of us would have to pay that sum today if we wanted to guarantee the solvency of our entitlement system for future generations.

Let’s explore this $88.2 trillion in a bit more detail.

We can divvy this liability into four parts. The largest is the $29.8 trillion needed to fund Medicare Part A, which covers hospital stays. Another $27.7 trillion comes from Medicare Part B, which covers doctors’ services. And $17.1 trillion stems from Medicare Part D, the prescription drug benefit that took effect in January of last year. The remaining $13.6 trillion comes from Social Security. What is interesting is the smallest of the four parts is the most debated in Washington. It is yet another example of the old rule that the amount of time spent debating a budget issue in Washington is always inversely proportionate to its cost.

When people think about these kinds of issues, they usually assume Social Security is the big problem. As these figures show, the unfunded liability from Medicare Part D alone—the new drug benefit—is greater than the entire Social Security shortfall. Taken together, Medicare’s unfunded liabilities are more than five times Social Security’s.

The total unfunded liability from Medicare and Social Security encompasses about 7.6 percent of U.S. GDP from here to eternity, which works out to 70 percent of all federal income tax revenues from here to eternity. So instead of paying $292,000 per person now, we could permanently sequester 70 percent of all current and future income tax revenue for use only on Social Security and Medicare. Or we could permanently raise income tax rates by 70 percent to accomplish the same thing—although we’d actually need to jack it up even higher because a large tax hike would probably discourage some people from working.

To save promised benefits, we would have to dramatically cut spending starting right now or raise income taxes and never bring them back down. And by doing so, we would only be covering the shortfall from Social Security and Medicare payroll tax receipts. All other existing sources of entitlement funding, including payroll tax revenue, copays, deductibles and premiums, would have to remain in place.

This is not a pretty picture. And as bad as the situation currently is, the necessary response becomes ever more drastic the longer we wait. If past is prologue, the most likely response may be to amend the current system—for example, by raising the retirement age or making the payroll tax more progressive. Many options would improve the fiscal fitness of our entitlement system and reduce the need for drastic action elsewhere in the federal budget. But let’s be honest. These remedies work only because some people would get less than they are currently slated to receive. Painful as that may be, the question is whether other options would be even more difficult.

At face value, fiscal policy may not seem a concern for the Federal Reserve. After all, Congress holds the power of the purse. But the Fed cannot be an indifferent bystander to the overall thrust of fiscal policy. The reason is straightforward: Bad fiscal policy creates pressure for bad monetary policy. When fiscal policy gets out of whack, monetary authorities face pressure to monetize the debt, a cardinal sin in my mind.

The Fed is not the answer to our fiscal woes. Congress, as keeper of the government’s purse and the sole body with the power to tax and spend the people’s money, and the president, who approves their spending, are where the buck should stop. But here is the rub: Voters like you elect the Congress and the president. History may place blame on this or that president or on Congress for failing to act. Ultimately, though, the responsibility for solving this looming fiscal issue rests with you, the voter.

You may remember my mention of Josiah Stamp earlier. If you don’t figure out a way to get your elected representatives to come to grips with the overwhelming problem of Medicare, your heirs may well end up like Josiah Stamp’s. Stamp refused to evacuate his stately home during Hitler’s bombings of London. He and his eldest son, Wilfred, were killed by a bomb in 1941. Well, under prevailing British law, in the event one could not determine who in the line of succession died first, it would be presumed that the eldest did. Thus, legally, Wilfred momentarily inherited his father’s peerage. Despite dying together, the Treasury levied the estate tax twice: once on the occasion of Papa Stamp’s death, then again immediately afterward upon Wilfred Stamp’s death.

If you don’t get your leaders to focus on solutions that cover the unfunded liability of these entitlement programs, you will be faced with a Hobson’s choice between a Federal Reserve that reneges on its most solemn duty and government tax measures far more drastic than ever occurred to the hapless Stamp family.

The Fed will not monetize our government’s debts. So you are left with the people you elect to represent you in Washington. In the end, that means you must turn to the person you look at every day in the mirror—you.

This is not a case where time heals all wounds. Indeed, it is the exact opposite. Time, in the case of our long-term unfunded liabilities, wounds all heels. And you and your children and your children’s children will be the heels who are wounded unless you demand that something be done about it.

Well, Luther, I am not sure that happy ending will put a spring in your step as you retreat to luncheon. Here is the point: Despite the constant changes occurring in the way the world is economically geared and measured, despite the imperfection of the data, and despite the ebb and flow of political tides, the Federal Reserve does its level best to get things right.

When I was invited to enter what some call “The Temple” of the Fed, Chairman Greenspan sat me down and told me that I only have one duty here and that is to pursue the truth. It is a great privilege to work for an institution that is so true to its purpose.

And Luther, maybe, just maybe, when all is said and done, my colleagues and I will be able to execute our mission with the same accomplishment and grace and humility with which you and Teresa have built your legacy in Fort Worth and in the investment community at large.

Thank you.