Academic journal article
By Wadhani, Sushil B.
National Institute Economic Review , No. 167
This article is primarily concerned with whether the US equity market is appropriately priced to reflect current global economic risks. From a policy perspective, this is an important question because if the US equity market were to experience a significant derating, it could cause a nasty recession.
Section 2.1 starts with a deliberately oversimplified Dividend Discount Model (DDM, hereafter), and shows that, under certain standard assumptions, the US equity market could be egregiously overvalued. We then consider alternative assumptions in turn.
First, we look more closely at our assumptions about dividends/earnings. Section 2.1 incorporates the effect of stock buybacks, while the possibility of a temporary increase in earnings growth is allowed for in section 2.2. Relatedly, we discuss various aspects of the 'New Economy' view in section 3. We consider whether recessions are less likely (3.1), and also whether recent technological developments can justify a permanent increase in earnings growth (3.2). Next, we discuss the optimism of Wall Street analysts (3.3), and also consider the sustainability of the current gap between the return and cost of capital (3.4), in trying to decide whether it is appropriate to assume a permanent increase in earnings growth.
We then turn our attention to the equity risk premium (ERP, hereafter). We initially examine what economic theory tells us about it (section 4.1), and then consider the effects of lower inflation on the ERP (4.2). We subsequently discuss whether stocks have been chronically undervalued in history (4.3 and 4.4) and whether one can justify a very low ERP (4.5). Our 'best guess' estimate of the current ERP is presented in section 4.6, and we contrast it with current return expectations in 4.7.
Section 5.1 discusses the potential role of the US stock market in exacerbating a hard landing, while 5.2 raises the possibility that a 'soft landing' of the US economy in 1999 could be associated with a further decline in the ERP. Section 5.3 considers the question of whether central banks should attempt to influence the level of equity prices. Finally, some conclusions may be found in section 6.
2. Are US equities overvalued?
With the US equity market having fallen initially nearly 20 per cent from its highs in August-September 1998, and then largely recovered its losses, it might seem strange that we are worrying about the potential downside. Chart 1 reminds us that in the crash of 1929, US shares fell initially about 40 per cent, then rose by about 44 per cent into early 1930 (for a net fall of 13 1/2 per cent as the Fed significantly cut the discount rate), before falling over 80 per cent over the subsequent two years. The bottom in 1932 was over 85% below the 1929 peak. On the other hand, several professional stock market strategists on Wall Street remain optimistic about the equity market. For example, Goldman Sachs' Abby Cohen says that the US equity market is likely to go on up.
2.1 A conventional formulation
For simplicity, we begin with a steady-state version of the dividend discount model (DDM), i.e. Gordon's (1962) growth model. We use this model as it is a commonly used workhorse in the financial community, and is a convenient vehicle to illustrate some of the relevant arguments. We shall consider more sophisticated valuation models at a later stage.
The model asserts that:
DY + g = r + rp (1)
DY = Dividend yield g = Expected long-term, real growth rate of dividends r = Real interest rate rp = Equity Risk Premium
In practice, the variables in equation (1) are difficult to measure. We may initially proxy the variables as follows:
(i) Real interest rate, r:
We use the yield on US Treasury Inflation-Protected Securities (TIPS, hereafter), which is currently about 3.7 per cent.
(ii) Real growth rate of dividends, g:
One possible assumption is the actual, long-term growth rate of real dividends, which, over the 192697 period, is only about 1. …