decrease in purciiasing power because of inflation, and, oh yes, just as big a debt reduction in real terms for the Govemment.

But there are other well-recognized causes of inflation. Entidement programs and social welfare, set up in the 1930s under the New Deal and expanded many times on all governmental levels since the war, malce it very hard to reduce budget deficits. So does the ever-increasing national debt, now 25% of the GDP (gross domestic product). While we are seeing the first major attempts to reduce some entidement programs, it appears at the time of this writing that it will be a long, uphill battle. Conscientious efforts have been made to reduce the deficit, and indeed the first balanced budget since the sixties was projected in January 1998. Still, some cosdy new event that costs billions of dollars inevitably comes along to hinder the process. The S&L bailout, the Gulf War, Bosnia, Iraq, Somalia, Haiti, numerous natural disasters are all examples of this type of unbudgeted phenomenon. In sum, it is hard to get control of spending for a myriad of reasons, only a few of which are listed here. That means a balanced budget, or more unrealistically, a budget su lus over time, is going to be more difficult to maintain than many believe.

After World War II another cmcial new socioeconomic variable was introduced into the inflation picture in the United States-a variable that has been important ever since. In traditional microeconomic theory, labor is treated as a commodity, and its cost, like those of all other commodities, might rise or fall according to changing market conditions. In 1946 Congress passed the Full Employment Act, making labor something different. In the parlance of some economists, labor is an administered market-or a market in which the flexibility of price movement is restrained.

In the labor market from the end of the war to the early eighties, wages were expected to be renegotiated upward regardless of prosperity or recession. And to the early eighties, at least, to move up faster than the rate of inflation. Wages, then, were a major contributor to rising prices at that time.

Through the rest of the eighties to the present, however, higher unemployment and a free trade policy resulted in wage rates showing a significant slowdown, as many jobs-and at times entire industries- were exported abroad. These factors led to a drop in the percentage of American workers who were union members, reducing further the power to increase wages.

Even so, inflation has risen steadily at over 3%. Much of the increase came from the service part of the economy, now projected to be over

Enter the Second Horseman

Table 13-1 showed how the disparity between the rates of return of stocks and govemment debt spiked sharply with the much higher rates of inflation endemic to the postwar period. Table 13-2 is identical to Table 13-1 in format, but retums for stocks, bonds, and T-biUs are calculated using a 50% income tax rate.

Over the entire 1802 to 1996 period stocks provided a 5.9% annual retum after both inflation and taxes, almost triple the retum of T-bills and more than 2A times the retum of long govemment bonds. Again glance at the three major subperiods. As you can see, the disparity in retums grows larger as the years pass. As is evident with rising inflation and taxes, both bond and T-bill retums, positive in the preceding two

60% of GDP, and continuing to grow rapidly. The service segment has been estimated to have increased its pricing in high single-digit or low double-digit figures in recent years, accounting for a significant share of the continuing inflation.

Since 1980, two major contributors to price increases in the previous two decades-oil and commodities-have actually been declining, and in inflation-adjusted dollars are well below their peak levels of 15 years ago. Although there is still a reasonable supply of oil, the dynamics of oil pricing could change rapidly for political or economic reasons. The Saudi Arabian Royal Family, for example, is under increasing pressure from the countrys Islamic fundamentalists.

The Royal Family is clinging to power in the face of an increasing undercurrent favoring Islamic fundamentalist change. If such a change does occur, it is quite conceivable-with Saudi Arabia being by far the largest oil exporter in the Middle East-that the resulting shocks could send oil prices soaring again, and inflation with it. A price increase could also come from other totally unpredictable events that could hmit production or rejuvenate the OPEC Cartel.

Finally, commodity prices, after lows in the early 1990s, could again head higher, as could wage rates. Thus three major causes of inflation in the 1970s and eariy 1980s, which have been dormant as prices continued to rise in recent years, all appear to have the potential of rekindling. If this does take place, we could be faced with inflation at higher rates than the 3.5% ofthe past 10 years.

The bottom line is that inflation rates are likely to stay at curtent levels or increase in the years ahead. Which indicates the outlook for bonds, T-bills, and other debt continues to be mediocre.

All returns are averages for the period, expressed as percentages. "Nominal Yield" is the mean annual T-Bill return before infladon. "Infladon" is the average increase in the Consumer Price Index. Data: All data 1802-1945, Gold data 1802-1995: Jeremy Siegel.

All other data 1946-1996: Dreman Foundadon. - Tax rates: Income: 50% 1914-1987; 35% 1988-1996.

Capital Gains: 25% 1914-1996.

major subperiods, tum negative in the 1926 to 1996 period, while stocks-despite two enormous crashes in 1929 and 1987-still provide a 4.2% annual retum.

The postwar period really lights up the scoreboard. As Table 13-2 demonstrates, T-bill retums were negadve for the 50 years following 1945 after inflation and taxes, while bond retums were in the red for the first 34 years. Stocks outperformed bonds and T-bills in every single subperiod between 1945 and 1996. Even in the 1963 to 1979 period, when stocks had a negative annual retum of -2.3% after inflation and taxes, it was still better than the .4% of bonds or -2.7% of T-bills. Gold, for the sha -eyed reader, did better during this time, providing an average retum of 10.9% annually, but for the whole 1945 to 1996 period it was in the red, -0.13% a year.

The outperformance of stocks, which was significant in the 1802 to 1870 period, when inflation was a nonevent and income taxes were al-

Table 13-2

Annual Returns to Stocks, Long-Term Government Bonds, T-Bills and

Gold: Adjusted for Inflation and Taxes

1802-1996