On February 4, 1994, Federal Reserve Chairman Alan Greenspan announced a quarter-point rise in federal funds rate, which is the overnight interbank lending rate and a basic instrument of monetary policy. It was the first of four interest rate hikes. By late May, the Federal Reserve had driven up short-term interest rates by a percentage point and a quarter.
From the beginning, things went badly. Contrary to Greenspan's professed purposes, long-term interest rates soared. Thirty-year fixed-rate home mortgage loans had been available at about 7 percent before the Federal Reserve acted. At this writing they are over 8 1/2 percent. Some medium- and long-term rates actually rose by more than short rates at first: three-year notes jumped a full point and a half; 10-year bonds jumped a point and a quarter before the Board's fourth rate-hike, of a half-point, in May.
Four months later, many questions remain. Why did the Federal Reserve act? What went wrong? What can we do? In answering these questions, we will find a deep incoherence at the heart of Federal Reserve decision-making. The episode suggests a clear need, not just for policy change, but for institutional reform.
The Board's Self-Criticism
Neither economic theory nor history dictate that long-term interest rates necessarily rise when short rates rise. It depends on circumstances. In this case the Federal Reserve's own documents reveal how badly it misjudged what the real circumstances were.
On February 23, one of the Board's technical experts, Vincent Reinhart of the Division of Monetary Affairs, delivered a memorandum to Donald Kohn, Director of the Division, that helps explain the misjudgment. Reinhart's retrospective memo distinguishes three hypothetical cases:
First case: A tightening [of short-term rates] that whittles down inflationary expectations, say by convincing market participants that the central bank is willing to act on its distaste for inflation, will permit nominal rates to be lower in the future . . . The policy move might result in very little, if any, rise in long-term rates. This, more or less, was Greenspan's hope.
Second case: Tightening that is deemed to be insufficient to trim future inflation, say because it was viewed as only keeping up with an inflation impetus, brings with it no expectation of lower future nominal rates. . . . As a result, more of the increase in the short rate would pass through to the value of long rates. This would have been an intermediate case, short of success, but better than failure.
Third (and relevant) case: "A tightening that raises inflation fears, say, because it was viewed as suggesting future, previously unexpected pressures on inflation that would only be incompletely offset by the policy action brings with it the expectation of higher future nominal interest rates. In such circumstances, the current short rate would be accompanied by an upwardly revised outlook for short rates over the long haul. . . . As a result, the increase in the short rate would have a magnified impact on long rates." This is what actually happened.
In other words, by the analysis of the Federal Reserve's own expert staff, the rise in long rates that followed suggests that the move on February 4 was a mistake. To be precise, it was neither necessary, because the inflation expectations with which it dealt did not exist beforehand; nor was it sufficient, because the move itself was too small to deal with whatever inflation expectations it itself created. Considering the source, this is a devastating indictment.
Thoughts of Chairman Greenspan
Why did the Federal Reserve take this action? Dutiful journalists reported that the policymaking Open Market Committee was out to sustain the recovery,"to engineer a soft landing," to "nip inflation expectations in the bud." Perhaps some members of the Committee actually believed this. But the record, notably the Federal Reserve's own official account of its policy, suggests that they had no serious basis for holding this or any similar view.
There is only one authoritative source: the semiannual Report of the Federal Reserve to Congress, mandated under the Humphrey-Hawkins Full Employment Act of 1978. Chairman Greenspan delivered such a report on February 22, 1994, nearly three weeks after the fateful February 4 rate increase. Yet this document conspicuously fails to justify the policy change on inflation-fighting grounds:
Greenspan's Point 1: There is no inflation problem. Greenspan begins by conceding that his previous fretting over inflation, expressed in his immediately preceding report in July of 1993, had proven unfounded: "On the inflation front, the deterioration evident in some indicators in the first half of 1993 proved transitory." Further, he reports, consumer price inflation in 1993 was the lowest since 1986. While falling oil prices contributed to this reduction, inflation was falling even excluding oil. What about commodity prices? Here Greenspan is explicit: commodity prices had "firmed" in the months before the February move. Yet "in the past such price data have often been an indication more of strength in new orders and activity than a precursor of rising inflation throughout the economy. In the current period, overall cost and price pressures still appear to remain damped." Greenspan goes on to note, correctly, that wage pressures remain absent -- and wages are a far larger share of costs than commodity prices.
Greenspan's Point 2: Growth is not excessive. Greenspan discounts the concern that the economy is growing at a rate that poses inflationary risk. "It is too early to judge the degree of underlying economic strength in the early months of 1994." This cautious opinion was vindicated in late April, when preliminary first-quarter 1994 real GDP growth was shown to have sharply decelerated from the high growth rate at the end of 1993. Greenspan does note that "attempts to force-feed the economy past its potential have led in the past to rising inflation," which is true enough. But Greenspan does not suggest that the economy is presently at, or even near, its potential.
Greenspan's Point 3: Inflation expectations have been falling. Here Greenspan turns to an explicit consideration of inflation expectations and their role in the inflation process. But inflation expectations have not been rising, by his account. To the contrary, Greenspan finds that both fiscal and monetary policy "have contributed to the decline in inflation expectations in recent years along with decreases in long-term interest rates." In particular, "the actions taken last year to reduce the federal budget deficit have been instrumental in this regard."
Why, then, raise interest rates? The key paragraph in Greenspan's testimony gives but one solid clue: "As the Federal Open Market Committee surveyed the evidence at its February 4 meeting, a consensus developed that the balance of risks had, in fact, shifted. . . . Real short rates close to zero appeared to pose an unacceptable risk of engendering future problems."
By "real short rates close to zero," Greenspan means the difference between a short-term interest rate such as the overnight federal funds rate, 3.0 percent on February 4, and the current rate of inflation, 1.9 percent for the three months ending January 1994. Since short-term rates are also what is now paid on very liquid bank deposits -- the near-equivalent of cash -- Greenspan's argument amounts to a call for a significantly positive real rate of return on money.
But there is a big problem with this argument: as a matter of history, it is untrue. Checking accounts and other cash equivalents earned no interest at all until the early 1970s -- there simply was no return on money. Hence, for all of postwar history up to that point, the real short-term interest rate on such holdings -- defined as the nominal interest minus inflation -- was actually negative. It was zero minus the rate of inflation. Yet despite this, inflation remained quite low, by more recent standards, from 1947 through 1970. If we look at slightly less (but still very) liquid assets, such as loans in the overnight market for bank reserves (federal funds market), we find that real interest rates in this market averaged only 1.6 percent from 1960 to 1972. And with steady growth and high employment, the sixties were the best period of postwar macroeconomic performance.
In Search of Theory
Ah, but isn't it the Federal Reserve's responsibility to look ahead and prevent inflation before the symptoms appear? And can't we accept Alan Greenspan's argument that the moves in February and March represent low-cost insurance to stave off inflation? Hardly. To "look ahead" for inflation or anything else in economics, one needs a theory, something that will predict the inflation that the data does not yet reveal. And to know whether insurance is a good bargain, you need to know the risks. Yet the Federal Reserve itself rejects the only economic theory that predicts inflation now.
That theory is monetarism. Some monetarists, such as Professor Allan Meltzer of Carnegie-Mellon, have been calling loudly for tighter policy, on the claim that rapid growth in some measures of the money supply signals higher inflation. But the statistical record of this idea since 1983, when the Federal Reserve abandoned its last fling with monetarism, is extremely poor. Recognizing this, Greenspan's own testimony explicitly rejects monetarist arguments: "We will continue to monitor developments in money and credit, but in 1994 as in 1993 the FOMC [Federal Open Market Committee] is unlikely to be able to put a great deal of weight on the behavior of these aggregates [that is, the indicators of money supply] relative to their ranges." Any present confluence between monetarist theorizing and monetary policies is, we are told, accidental.
If he is not a monetarist, is Greenspan perhaps a closet Keynesian? Traditional Keynesians argue that inflation threatens when unemployment falls too low, when capacity limits are reached, or in some variations when economic growth is too rapid. But while Greenspan does carefully assess each of these threats, he then explicitly dismisses them -- and rightly so. The data, showing unemployment at 6.5 percent, capacity utilization at 83.6 percent, and growth averaging perhaps 3 percent (the exceptional strength of the fourth quarter of 1993 having faded), indicate that the economy remains below past inflation threshholds. Non-traditional Keynesians (such as myself) would also argue that severe competitive pressures on U.S. industries -- rooted in rising trade with low-wage countries, a decline in labor standards, and fire sales of commodities from former communist countries -- will keep a lid on inflation, even if employment levels rise. Thus Keynesianism cannot explain Greenspan's actions, any more than monetarism can.
Is Greenspan then a disciple of the hard-line "new classical economics"? Conservative commentators have long rumbled about the need for the Federal Reserve to adopt a zero inflation targetþa "monetary rule," in effect, of tightening policy whenever the inflation rate is positive and political circumstances permit. Legislation imposing such a target, and thereby annulling the formal goal of "full employment" embodied in the 1978 Humphrey-Hawkins Act, has been proposed in Congress and has had supporters on the Open Market Committee. The idea is that the economy will fluctuate naturally around an equilibrium rate of unemployment, irrespective of policy; the Federal Reserve should concern itself only with stabilizing the price level, by taking a super-tough stance to subdue inflation expectations.
Again, the description doesn't quite fit. For one thing, the new classical theory has a monetarist core. In this line of argument, achieving price stability requires the Federal Reserve to focus on control of the money supply, which it is not doing. Moreover, if inflation expectations are rising, why would Greenspan explicitly state that they have been falling? There is no new classical coup at the Federal Reserve Board.
Is the Federal Reserve then reacting to the rising price of gold? This would make the Board into a neo-supply-side institution and a secret ally of Jude Wanniski, who has called (on the op-ed page of the Wall Street Journal) for tighter policies to stabilize the price of gold. Greenspan himself has an embarrassing past as a gold bug, and his testimony nods in this direction, stating that the price of gold "has been especially sensitive to inflation concerns." But one subordinate clause seems a thin reed on which to base a theory of Federal Reserve action.
The most intriguing possibility -- based on Greenspan's obscure reference to the "unacceptable risks" of low interest rates -- is that he has embraced some antique theoretical ideas, traceable to the early twentieth century Swedish economist Knut Wicksell. These ideas, proposing a "natural interest rate," concern the relationship between the optimal rate of interest, the physical productivity of capital investment, and the rate of growth. According to this doctrine, interest rates should rise when they fall below the "natural" rate of return on capital, which in simple steady-state models is equal to the long-run sustainable real rate of economic growth. If this rate is, say, 2.5 percent, and inflation is 2 percent, then such a rule would indicate a target level of 4.5 percent for nominal interest rates.
A Reuters dispatch from early May lends some support for this view, noting that that Federal Reserve officials "want to shift monetary policy away from being accommodative . . . to a neutral stance that neither acts as a spur nor a drag on growth." Reuters quoted Federal Reserve Board Governor Lawrence Lindsey: "We're not quite in neutral. But we're substantially closer to it."
Lindsey and Greenspan may be using "neutral" where they mean "natural" in the Wicksell sense. The confusion of words could be an appeal to metaphor, namely to the neutrality achieved by shifting an automobile's engine out of gear, in order to explain this idea to the vulgar.
But even if we get the terminology straight, applying the doctrine of a natural interest rate presents three major problems. First, we have no measure of the natural real rate of return on capital (indeed, we have no coherent measure of the capital stock); the model only works in a highly simplified theoretical setting. Second, the long-run sustainable rate of growth is only relevant after the economy is at full employment. So long as unemployment remains high, the economy can grow faster than its present rate, with lower interest rates helping it along. Hence, this theory does not suggest that the best interest rate now is higher than it was in January. Finally, this theory does not tell us why the short-term interest rate should be adjusted to the "natural" rate. To the contrary, it would seem that long-term bond rates are much more relevant to the problems that interested Wicksell, and real long-term rates remain extremely high.
- John Maynard Keynes long ago summed up the arguments in this field:
- I am now no longer of the opinion that the concept of the 'natural' rate of interest, which previously seemed to me a most promising idea, has anything very useful or significant to add to our analysis. It is merely the rate of interest which will preserve the status quo; and, in general, we have no predominant interest in the status quo as such.
Apparently, others have accepted that verdict. In the last 50 years natural interest rate theory has hardly developed; many textbooks don't even mention it. It would be amusing, but strange, if Greenspan and Lindsey were intent on reviving this archaic doctrine.
There is the possibility that Greenspan was acting to defend the dollar. Press reports suggest that Japanese money has been flowing out of foreign investments, which would weaken the dollar against the yen. Perhaps Greenspan felt higher interest rates would stop this trend. If true, this motivation has three problems. First, it's dumb, since raising U.S. interest rates can only deepen the financial crisis in Japan by drawing funds back out of Japan. Second, it won't work (and didn't: the dollar didn't stabilize). And third, it's a total secret so far as reporting to Congress and the public is concerned. The words "dollar," "international," and "Japan" do not appear in the February Report, and the phrase "exchange rate" appears only once, unimportantly.
Can some non-economic, or at least non-macroeconomic, theory explain the rate hike? Perhaps Greenspan was trying to set a course towards tighter money policy before Clinton's two new appointees to the Board, Alan Blinder and Janet Yellen, come and try to stop him. Or perhaps Greenspan, alongside the overwhelming Republican majority on the Open Market Committee, has simply decided to torpedo the Clinton administration's economic recovery, and with it the immense progress made in 1993 in budget deficit reduction.
Each of these political motivations, if true, would constitute grounds for impeachment, not only for Greenspan but for all who voted with him on the Federal Reserve Board. It is not the Federal Reserve's job to regulate the outcome of elections. Of course, hard evidence on these questions is lacking and, in the final analysis, unlikely to surface.
One last possibility came to light when Greenspan returned to the Senate on May 27 to defend his policy. Perhaps he felt a need to pop a bubble in the stock and bond markets, and specifically to help drive stock and bond mutual fund investors back to the eroding deposit base of the commercial banks. Greenspan was astonishingly explicit about this to the Senate, stating that small investors had become accustomed to "an unsustainable combination of high returns and low volatility;" that they failed to appreciate the inherent risks of holding long-term assets. In reality, of course, these high returns were only unsustainable after February 4 -- because Federal Reserve policy had changed.
The crass clientelism of this motive gives it, one fears, the awful ring of truth. The high spreads between long and short rates in recent years have been a thinly disguised bailout for commercial bank profits. So why not now bail out commercial bank market share, by forcing small investors out of long-term investments and back to the security (and low-interest rates) of insured bank deposits? The mind wrestles with the anti-social implications of this policy, as well as with the effrontery of admitting as much, even in obscure language, to the Senate. If a tie-dyed Marxist had made the accusation, few would believe it. Absent formal inquiry into the FOMC transcripts -- which is called for -- I have trouble believing, even now, that a bank bailout was a key motivation before the fact.
We are left, then, with the thought that the Federal Reserve Board does not know what it is doing. This is the "Wizard of Oz" theory, in which we pull away the curtains only to find an old man with a wrinkled face, playing with lights and loudspeakers. *(See Footnote) Absent stronger evidence from Greenspan's mouth or the Board's files, I fear that this is the conclusion on which we must, for the moment, rest our case.
In this view, Federal Reserve policy reduces to a syllogism: "good times are followed by bad times; therefore good times must be prevented." An old chestnut attributed to William McChesney Martin, who was Federal Reserve Board Chairman in the 1960s, holds that the Federal Reserve's job is to "take away the punchbowl just when the party gets going." Has this feeble cliche, with its stupid implications, become the central basis on which monetary policy is now made?
A Breach of Faith
We arrive at the view that a group of nine men and one woman, nine Republicans and one Democrat, decided to raise interest rates because on balance they felt, for no particularly coherent reason, that raising interest rates was the thing to do. Should we, the public, the Congress, and the Administration, accept this?
Clearly not. The actions so far have not led to recession. But the costs are nevertheless substantial -- a huge increase in the "interest tax" paid by consumers and businesses to their banks, as well as slower growth and higher joblessness.
To take a typical case -- my own -- a rise of 1.25 percentage points in the short-term interest rate will translate into an additional $2,000 in interest payments, next year, on my $160,000 adjustable-rate mortgage. This is a tax increase many times as great as last year's increase in the income tax. And I am paying this tax to a private commercial bank whose shareholders are no doubt on average richer than myself.
For the Administration and Congress, there is another, equally serious consideration. With the public holding $3 trillion of federal debt, each one-point increase in interest rates eventually imposes $30 billion in net new annual interest costs on the budget, while slower growth cuts revenues and adds to spending for unemployment insurance and welfare. This is enough to upset, in a radical way, progress made toward lower budget deficits in 1993.
- The 1994 Economic Report of the President, transmitted on the same day as the new Federal Reserve policy (February 4), reveals clearly how much the Clinton administration and Congress were relying on lower, not higher interest rates:
- Long-term nominal and real interest rates dropped sharply in 1993. The decline in rates was closely linked to the proposal and enactment of the Administration's budget . . . Lower Federal borrowing reduces interest rates directly, by reducing demand for credit. [And] a more prudent fiscal policy reduces the likelihood that the Federal Reserve will need to pursue a restrictive monetary policy, and so reduces expected future short-term rates. (p. 35).
- Later the report notes:
- "With a greatly improved outlook for the Federal budget deficit, the Council of Economic Advisers expects long-term interest rates to remain relatively low for the foreseeable future -- which will help keep economic growth on track. Low interest rates are the key ingredient that should allow the economy to grow in the face of future large deficit reductions, which would otherwise tend to contract the economy."
(p. 56)
Thus, higher interest rates cut twice against the administration's best hopes for sustained growth. They slow economic growth directly by deterring job creation, and they raise the deficit itself, which in turn increases political pressure for an even more contractionary fiscal policy.
Toward Accountability
In 1993, Alan Greenspan and the Federal Reserve Board encouraged deficit-cutting, leaving the impression that deficit reduction was the key to low interest rates. President Clinton repeatedly made the connection between the two. Members of Congress repeatedly justified tough votes on taxes and spending to their constituents in the same terms. Unfortunately, the administration did not obtain an explicit commitment that the Federal Reserve would reward progress on the budget deficit by maintaining low interest rates. The apparently coordinated economic policy of the administration and Congress with the Federal Reserve in 1993 has been rejected by the latter, without the consent of the former. The cumulative effect raises the deepest suspicions of bad faith on the part of the Federal Reserve Board.
The immediate problem is bad policy. This Congress can change, if it chooses. The Federal Reserve is not, under the Constitution, a fourth branch of government. It is an agency created under the Constitutional provision that Congress has the power to "coin money and regulate the value thereof." Congress wrote the Federal Reserve Act, and can change it. It has already done so in a general (if ineffective) way, for example, by writing the objectives of "maximum employment, production and purchasing power" into the 1946 Employment Act and "full employment" alongside "reasonable price stability" into the 1978 Humphrey-Hawkins Full Employment and Balanced Growth Act.
To pass a law on current policy, however, would involve the President directly in monetary policymaking, and would thus end the independence of the central bank from the executive branch. It seems a draconian step to remedy a short-term problem of policy incoherence.
Alternatively, Congress could put a monetary directive into a concurrent resolution. Concurrent resolutions, such as the annual budget resolution, do not require the President's signature and thus lack the force of law. But past Federal Reserve chairmen have acknowledged that such resolutions are binding in effect, because the Federal Reserve Board is a creature of Congress.
In 1975, in House Concurrent Resolution 133, Congress directed the FRB to conduct policy "so as to lower long-term interest rates." At the end of 1982, Congress ordered that monetary policy "achieve and maintain a level of interest rates low enough to generate significant economic growth and thereby reduce the current intolerable level of unemployment."
Last April 28, 44 members of Congress wrote Chairman Greenspan asking that interest rates be raised no further. This request was ignored, and the time is ripe for stronger action. A directive ordering a gradual roll-back of interest rates -- a shocking thought to many -- is distinctly thinkable today.
Shedding Light
Over the long haul, it is neither practical nor wise for Congress to micro-manage monetary policy. Yet experience also tells us that the 19-year effort to impose rationality and coherence on the Federal Reserve solely through semi-annual reporting requirements has failed. It's time for something more effective.
One proposal, by Congressman Lee Hamilton and Senator Paul Sarbanes would abolish the Federal Open Market Committee. At present the FOMC is composed of the seven Federal Reserve Board members and a rotating group of regional Reserve Bank presidents. The Hamilton-Sarbanes plan would reconstitute the twelve regional Federal Reserve Bank Presidents as a non-voting Open Market Advisory Committee. Alternatively, Representative Henry Gonzalez proposes to make the heads of the regional banks Presidential appointees. Either measure would constitute an improvement over the present system. The Gonzalez proposal would also permit the President to bring gender and race diversity to the regional Federal Reserve bank presidencies for the first time. The antiquated structure of regional Federal Reserve banks (there are two in Missouri and one in California) as well as of the Open Market Committee (which gives preferred voting rights to New York, Chicago, and Cleveland over other districts) is also ripe for change. But these important reforms do not really address the basic issue of policy coherence.
What would help most would be an immediate end to the reign of secrecy in Federal Reserve decision-making. If we knew in detail and in real time how and why each member voted on each decision, we would not assure a coherent or rational result. But at least we would know the basis for monetary policymaking -- if there is one. And we could take the next step, which is to evaluate the coherence of the Federal Reserve Board and Open Market Committee, now carefully shielded from public view.
In the one positive development this year, the Federal Reserve has itself stripped away all justifications for the deep secrecy surrounding its monetary policy decisions. Until February 4th, the Federal Open Market Committee indulged in a charade, whose only effect was to reduce the Federal Reserve's political visibility: The FOMC issued a policy directive, but released it to the public only six weeks later, after the subsequent FOMC meeting. Money markets, of course, could deduce the new policy within hours, from changes in the pattern of Federal Reserve operations. But the public and Congress had no official statement, and no explanation, on which to hang a discussion of the policy merits.
On February 4, and again in the following weeks, the FOMC dropped the charade. Greenspan simply announced that the federal funds rate would rise by a quarter-point -- an immediate disclosure of the FOMC directive. What followed was exactly what advocates of immediate policy disclosure had long predicted. The federal funds rate adjusted, immediately, to the announced target. This happened before any open market operations were conducted. And the Federal Reserve had no difficulty maintaining the federal funds rate exactly where it wanted it, thereafter. This tells us that all previous arguments for secrecy surrounding the FOMC's directive were captious, as critics had argued. Why not then take the next step, and open the FOMC meeting at which the directive is decided?
Those who defend FOMC secrecy cannot argue, as they have done up to now, that real-time public scrutiny of Federal Reserve decision-making would affect implementation of monetary policy. We have put that proposition to test, and found that monetary policy is easier to implement, and more effective, when the decision is publicly known. We are left only with arguments that somehow spontaneity, or something, would be lost if FOMC proceedings were public.
But if the Federal Reserve cannot give us a coherent account of its policies, why should we take it on faith that wonderful things happen behind closed doors at its meetings? Members of Congress, particularly more senior members, have direct experience with moves toward greater openness, both in the reforms of 1975, which opened up the Committee processes, and in the creation of C-SPAN in the 1980s. Some congressmen cannot function without a script. Others function very well, and operate as comfortably in public view as behind closed doors, by mastering the details of the rules and of the legislation before them. Of the two types, which do you suppose is more effective and commands more respect?
Congress discovered that there was nothing, intelligence appropriations and a few other things excepted, in the making of public laws that could not be shown to the public. In the case of Federal Reserve monetary policy (bank regulatory matters are something else), there is even less to hide. Why are economic data, the runs of models, and the arguments of monetarists and Keynesians, considered a state secret?
President Clinton has recently appointed two first-rate economists to the Federal Reserve Board: Alan Blinder of his own Council of Economic Advisers and Janet Yellen of Berkeley. Neither would have any difficulty making a public case for a more rational, better defended policy. It would be interesting to see whether any of the old guard could effectively challenge this high-powered pair in a public setting. And it would be a shame if their voices were lost behind the veil of obscurity that now surrounds Federal Reserve decisions.
Congressman Gonzalez, chairman of the House Banking Committee, has suggested that a camera be placed in every Federal Open Market Committee meeting. Fed-SPAN -- that's the ticket. Blinder and Yellen should team with Lindsey (no intellectual slouch, though conservative) and simply put this issue to the Board. If they won't, or if the Board won't go along, then let Congress do it for them.
* The parallel is not accidental, as scholarship has shown that "Oz" was a monetary parable. Specifically, the Emerald City was Washington (where greenbacks were made), the tin man (rusted solid) was a depressed industrial worker, the Scarecrow a depressed farmer, and Dorothy wore silver slippers on the yellow brick road. William Jennings Bryan is the Cowardly Lion; the wizard was the President, dispatched in the end in a hot air balloon. The parable was elucidated by Henry Littlefield, "The Wizard of Oz: A Parable of Populism," in the American Quarterly of Spring, 1964. I am indebted to Lowell Dyson of the Economic Research Service, USDA, and to Jim Devine and Michael Genovese of Loyola Marymount College for posting information on the Internet on this topic. (back to text)
