WASHINGTON, Nov. 3 (UPI) -- Edwin Truman, former head of the International Finance Division of the Fed, spoke at the Institute for International Economics Thursday about the need for inflation targeting in monetary policy. However, it wouldn't have worked in the late 90s, because published inflation figures ignore asset prices. Maybe the real problem of monetary policy is the lack of any functioning compass at all.
Truman's thesis, published in "Inflation Targeting" (Institute for International Economics) is that by targeting inflation, using a band with a lower and upper bound, central banks can increase the transparency of their monetary policy. If inflation exceeds the target band, they tighten, if it falls below, they loosen, and if it is within the band, they are neutral.
New Zealand was the first country to target inflation, in 1989, using a 0 percent to 2 percent band, and reducing the salary of the Governor of the Bank of New Zealand when inflation rose above 2 percent. Today, 22 countries have adopted inflation targeting, some of which are within the target band, and others above the band but with inflation reducing towards it. In practice, according to Truman, an inflation target that includes zero within its target band has the danger of becoming too contractionary and leading to deflation; nevertheless, several of the inflation targeting countries still include zero within their target band. Truman believes that the United States, the European Union and Japan should formally adopt an inflation targeting policy, in Japan's case expanding money supply so that consumer prices cease to decline and begin a slow inflation.
The problem with inflation targeting, however, is that it fails to take account of asset prices. Before 1980, U.S. house prices, in the form of mortgage interest payments, were included in the consumer price index, but house price bubbles in 1979-80, 1988-89 and 1999-2003 all proved inconvenient, so housing was effectively removed from the index by the Bureau of Labor Statistics, who invented an "owners equivalent rent of primary residence" which can be and has been manipulated to eliminate rapid house price increases when these would suggest inflation was rising too fast. In 2001-2002, house price increases contributed substantially to overall inflation, but their effect was largely dampened by the corresponding decline in the stock market. In 2003, of course, both house prices and the stock market have been rising, so true inflation is much higher than currently believed.
However, the asset price inflation that is of most concern is the stock price inflation of the late 1990s. When asked, Truman claimed that it was impossible to devise a sound methodology to include asset prices, in particular stock prices, in an inflation calculation. This is nonsense. As I discussed in an article some months ago, a number of studies have been done of the "wealth effect" of rising stock prices, and the consensus appears to be that around 4 percent of the rise in stock valuations leaks out into the economy in additional spending. Very well, then take 4 percent of the increase or decrease in value of common stocks, divide that figure by the year's gross domestic product, and add the resulting percentage to the consumer price index (preferably having first adjusted this to take true account of house price movements.)
For example in 1991 stocks increased in value by approximately $941 billion, 4 percent of which, the "stock price inflation," was $38 billion, while GDP was $5,986 billion. Stock price inflation was thus 0.63 percent of GDP. Add this figure to the official 3.06 percent CPI inflation figure, and the true inflation figure is 3.69 percent -- significantly less however when you include housing, which was deflating in 1991. This calculation, performed year by year, gives a true inflation index, including asset price inflation, which can be "inflation targeted."
The Fed may very well have believed that it was "inflation targeting" to a limited extent in the 1990s. Indeed, the historical record of the official "Consumer Price Index for all urban consumers" suggests that it was, with a target of 2 to 3 percent, a range that Truman recommended for the U.S. in his presentation. CPI inflation exceeded the target range three times during the decade -- 3.1 percent in 1991, 3.3 percent in 1996 and 3.4 percent in 1999. It fell beneath the target range once -- 1.7 percent in 1997. Pretty good inflation targeting, in other words.
Include stock prices, and the picture is very different. The excess above the range in 1991 was reduced by house price deflation, and stock price inflation in 1992-94 added only modestly to the reported CPI, so the true CPI remained below the 3 percent maximum of the targeted range. However, from 1995, it was a very different story. In that year, stock price inflation was 0.98 percent of GDP, so true inflation was not 2.54 percent, as reported, but 3.52 percent, well above the target range (house price inflation was positive in 1995, and accelerated as the decade wore on.) In 1996, alarm bells should have been ringing -- stock price inflation was 0.78 percent of GDP; add that to the reported 3.32 percent and you get 4.10 percent. Yet again it becomes clear that Fed Chairman Alan Greenspan should have tightened money sharply at the time of his "irrational exuberance" speech in December 1996.
In 1997, stock price inflation removed the apparent blip below the target range; its 1.31 percent added to the reported 1.70 percent gives you 3.01 percent, still above the target range although very close to it. In 1998, it's much worse; stock price inflation of 1.42 percent added to reported inflation of 2.68 percent gives true inflation of 4.10 percent, far above the official range (and house price inflation would have been adding substantially to the figure by 1998, too.) In 1999, finally, stock price inflation of 1.27 percent adds to reported inflation of 3.39 percent to give true inflation of 4.66 percent. Far from adding money at the end of 1998 and 1999, as he did, Greenspan should have been tightening sharply, had he been inflation targeting on a true inflation basis.
During 2003, stock prices have appreciated by approximately $2.5 trillion, 4 percent of which is $100 billion, 0.9 percent of 2003 GDP. In addition, house prices have risen by about $1.5 trillion during the year; using the 4 percent rule again adds another 0.6 percent to inflation. Since these changes have come in 10 months, their total of 1.5 percent should be annualized to 1.8 percent, at an annual rate. Add that to the 3.2 percent rate of inflation in the first 9 months of 2003, and you get total inflation of 5 percent. Certainly it is not deflation we have to worry about!
Monetarists may take comfort in the thought that measurements of money supply are not affected by asset price movements. M3 money supply increased by 9 percent per annum through the 1990s, but we now know that much of this was needed to fund stock price inflation and housing price inflation; money velocity did not in fact unexpectedly diminish, but inflation, when measured on a true basis, increased.
Monetarists have now expressed concern that money supply, which had continued to increase by 9 to 10 percent per annum throughout the recession of 2000-2003, suddenly stopped increasing in July 2003, to the extent that both M2 and M3 on October 30 were below their levels of 3 months earlier. At first sight, this looks like a very severe contraction, which can only bode ill for the stock market, but Wall Street's optimists have pointed out that part of the contraction results from retail investors moving funds from money market funds back into the stock market, as the market's recovery from its March 2003 lows was appreciated by the public.
There's just one snag to this. If this was true in 2003, then it must also have been true in 1996-99; in other words the already excessively high money supply increase, as measured by M3, of 9-10 percent per annum, was in fact still higher, and substantially so if money market fund movements in 2003 can wipe out a whole quarter's growth. Still more then does it appear that the Fed's policy in the late 1990s was utterly irresponsible.
My own view, for what it's worth, is that the modest rise in long term interest rates since June, and the consequent sharp deceleration in mortgage re-financings, are about to have a sharp effect on consumer spending, and thus on the economy as a whole. Third quarter GDP growth was artificially high at 7.2 percent, boosted by mortgage re-financings to June and by the tax rebates in July and August. Going forward, we can expect consumer spending to drop. We can also expect more of the true 5 percent per annum inflation rate to show up in the official figures. This in turn will have a huge effect on the Treasury bond market, and on interest rates generally -- to get a real rate of return of 2 percent (the historic minimum), if inflation is 5 percent, nominal interest rates must rise to 7 percent from their current 4.34 percent on the 10 year bond. This would be very bad news indeed for the housing industry, for the banking industry (which will lose both on its bond investments and through an increase in loan losses) and for the economy as a whole.
All this could have been avoided, if the Fed had adopted inflation targeting, based on true inflation, in the middle 1990s. Of course, if you don't trust the Fed to act responsibly in such a situation, because of political pressures, there is really only one solution: the fully automatic inflation targeting of a Gold Standard.
(The Bear's Lair is a weekly column that is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that, in the long '90s boom, the proportion of "sell" recommendations put out by Wall Street houses declined from 9 percent of all research reports to 1 percent and has only modestly rebounded since. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)
