April 3, 2000

The stock market bubble coincides with what has been termed "the immaculate expansion," meaning high growth with low inflation. The immaculate expansion is a puzzle for the traditional theory of aggregate supply. That theory states that as growth increases, the economy runs into capacity constraints, particularly in the labor market, resulting in higher inflation.

One interpretation of the immaculate expansion it that it proves that growth does not lead to inflation. In this view, the economists who believe in an aggregate supply curve are wrong.

Another interpretation of the immaculate expansion is that we are in a "new economy," characterized by accelerating productivity growth and global competition. Although the theory of aggregate supply may not be wrong, the "new economy" has increased aggregate supply. In this interpretation, the high valuation of the stock market reflects the favorable shifts in aggregate supply that result from technological innovation.

I want to propose a third interpretation of the immaculate expansion. I will suggest some ways in which higher stock prices create a favorable shift in aggregate supply. That is, the causality runs from higher stock prices to better macroeconomic performance, rather than the other way around.

Aggregate supply revolves around the relationships among prices, wages, and productivity. Other things equal, lower wages increase aggregate supply, by reducing production costs. Other things equal, higher productivity increases aggregate supply, also by reducing production costs. Finally, taking wages and productivity as given, lower price markups increase aggregate supply in that a lower markup reduces the price associated with a given amount of output.

For the stock market to affect aggregate supply, it must affect wages, prices, and productivity. There are at least five ways that it can do so.

1. Stock option effects.

Workers who receive stock options may accept lower wages than otherwise, particularly when the market is rising. Edward Yardeni (whose site is listed among my favorite links) thinks that this is important in the aggregate. I've always been a bit skeptical, in part because I do not think stock options cover a very large percentage of the work force. However, certainly in the high tech arena, which is increasingly important at the margin, stock options are helping to hold down wages. Otherwise, the market for the better web programmers might clear at a salary of over $150,000 per year.

2. Pension windfall effects.

Another point that Yardeni makes is that firms have not had to take money from operating revenues in order to fund defined-benefit pension plans. Instead, pension fund assets in the stock market have appreciated by more than enough to meet pension obligations. Yardeni's focus has been on how this affects measured saving in the economy. However, another result may be that firms under-estimate their labor costs. If their pension contributions have been zero, they may not impute a pension cost in their labor expense. This would lead to lower prices than would otherwise be the case, because a fall in imputed labor cost could have the same effect as a fall in actual labor cost.

3. Currency effects.

To the extent that the demand for U.S. equities reflects strong demand from overseas, an appreciation of the dollar will be part of the bubble. This in turn helps to hold down inflation, because it makes it possible for workers' real incomes to rise faster than their wage rates. The difference is the reduced cost of foreign goods.

4. Investment effects.

Higher investment leads to higher productivity. There is nothing artificial about this effect on aggregate supply. However, to the extent that higher investment is driven by the stock market, if the bubble were to pop, then this would drive down investment and productivity growth.

5. Internet subsidy effects.

Internet companies are losing money, with subsidies from shareholders. These subsidies enable them to offer products and services below cost (free personal computers, for example). This in turn puts competitive pressure on other companies in the same market, even if those companies do not enjoy "dot-com" status on Wall Street. The markup of price relative to labor cost is being driven to zero, and below.

All five channels of causality running from the stock market to the immaculate expansion are plausible. However, I do not know the magnitude of these effects. It could be that even if you add them all together, they reduce the inflation rate by less than one percent. In that case, the immaculate expansion remains largely a mystery.

On the other hand, in the late 1970's, when stocks were seriously undervalued, we had very unfavorable combinations of inflation and unemployment. This would suggest that the links between the stock market and aggregate supply could be strong.

When the bubble pops, we will get a better idea of how stock prices affect aggregate supply. If we truly are in a "new economy," then even when the bubble pops we will see low inflation. On the other hand, if an overvalued stock market artificially lowers inflation for a given level of unemployment, then when the bubble pops we could see the kind of stagflation that we saw in the 1970's.