Equity Markets and the UK Housing Market

This has been a turbulent year for stock-markets, with all the major markets down since the beginning of September. The charts here trace the S & P 500 index, the index of the leading 500 U.S. companies, the Nasdaq index (concentrated in newer U.S. high tech companies), and the FTSE-100 index of leading U.K. companies.

Why share prices have fallen

In the background to the share price decline lies a controversy, now raging for some years, over whether equities, particularly in the U.S., had become fundamentally overvalued. Virtually all the traditional valuation measures (such as share price/dividends, share price/earnings, the ratio of the stock market value of companies to the replacement cost of their assets and yield differentials between equities and bonds) seemed to be pointing towards the view that the U.S. market, in particular, had become overvalued. Empirical evidence supports the hypothesis of 'mean or trend reversion' implying that there is a tendency, both for very high, and very low share price levels not to be sustained but to 'correct' towards long-term levels. (see John Campbell's survey 'Asset pricing at the millenium', Journal of Finance, August 2000 for an outstanding exposition of modern research on asset, and particularly, equity, pricing, which confirms the trend correction phenomenon).

Many reasons have been advanced to explain the high equity values of the first four months of 2000. Analysts point to the longest U.S. peace-time expansion in a century, and to new technology boosting the underlying growth rate, making markets more flexible and so reducing inflation risk. Indeed, since the mid 1990s, there has been a notable increase in U.S. productivity growth. For example, output per head increases in manufacturing have recently been running at 5-6 percent per annum. The figures are probably flattered both relative to the past and relative to Europe by changes in measurement adopted by U.S. government statisticians, which make a considerably greater allowance for quality change (hence note the decline in the quality adjusted cost of computing). Past experience also teaches us that in booms, rates of factor utilisation always rise so that output per head rise more rapidly than proves sustainable in the long run. Nevertheless, even with these qualifications, it has hard to disagree with the basic proposition that productivity growth rates have increased. By the same token, growth of output is higher and inflation lower, though these figures are also flattered in the U.S. by more optimistic ways of measuring them.

Another factor is the particular skill shown by the U.S. Federal Reserve in avoiding inflation and promoting stability, as well as shifts to inflation targeting monetary policy regimes around the world. Indeed, many professional investors appear to believe that stock market valuations are effectively underwritten by the Federal Reserve in that, a rapid stock-market decline of the order of 20 to 25 percent, would result in a correspondingly sharp reduction of U.S. interest rates, stabilizing the market. If investors believe that they are protected by such an insurance policy, the average value of equities could indeed be permanently higher. Marcus Miller and colleagues at Warwick University claim that ownership of such an insurance policy (or 'put option') could explain 20 percent or more of the current level of U.S. equities. There is merit in all these arguments, but we belong to the sceptics who believe share prices were overvalued.

Thus, one explanation of the fall in share prices from the highs of the year is that it is part of the natural tendency of prices to revert to long-run trend. However, a second reason is that there are now serious concerns about the sustainability of high growth rates. Since the high equity valuations have depended, in part, on high anticipated earnings, associated with high expected growth rates, slower economic growth implies lower equity prices. One reason for slower growth is high oil prices, for months now at twice their average levels of the previous ten years, and even higher relative to 1999 levels: larger amounts of consumer and business cash are being siphoned off to the oil producers, lowering spending on non-oil products. Slower spending is also influenced by the decline in the dot.com investment boom. The risks of such dot.com ventures have become more apparent, given the 'winner-takes-all' nature of the competition in many dot.com market sectors. For the U.S., two related factors helping to explain the slowdown in growth are the overvalued exchange rate and the highest trade deficit relative to G.D.P. in U.S. history. The large trade deficit has been financed by foreign capital flows, by European companies and other private investors acquiring stakes in U.S. high-technology firms, especially in the telecom and dot.com sectors. However, another major factor has been the tendency by U.S. corporations to borrow in the European corporate bond market, which has increased greatly in size and depth since the advent of the EURO. Borrowing in EURO's and shifting the proceeds not invested in corporate activities in Europe into Dollars has been a substantial element in U.S. capital flows in the last two years.

The slowdown in U.S. growth can be regarded as a healthy corrective for the U.S. economy, bringing lower trade deficits and reducing internal inflationary pressure. It will alter the supply-demand balance in oil, eventually lowering oil prices, reducing inflationary pressure round the world. However, there is considerable uncertainty about this correction process. To add to the uncertainty, tensions in the Middle East raise the spectre of disruption to oil supplies.

These uncertainties have been reflected in the very high levels of volatility in share prices this year, particularly in technology stocks. A transition to lower growth rates from such a long economic expansion is almost bound to raise volatility. But the rapid pace of change of technology raises uncertainty about individual stocks greatly and not only in the dot.com sector discussed above. Many 'old economy' shares have gone through enormous price changes this year as major changes in the nature of retail competition have been taking place. Higher volatility necessarily implies lower share prices as investors seek safer returns.

The impact of lower share prices

It is likely that the stock markets may not yet have fully appreciated the nature of the 'feedbacks' currently at work. Lower share prices have the effect of reducing spending. Our own past research supports the consensus view, that, in the U.S. (and indeed the U.K.), 3.5 to 5 cents of every dollar of stock market wealth is spent by consumers each year, with spending lagging behind share price movements. The longest lags arise from the part of wealth tied up in pension funds lying outside the direct control of households: these tend to adjust contribution rates very slowly in response to changes in asset values. Business investment too is affected by equity prices, since lower equity prices raise the cost of capital. A sustained fall in share prices, therefore, leads to lower spending and lower income growth, which in turn, by lowering growth expectations can further lower share prices.

But there is yet more to this. The notion of the 'financial accelerator' has been used by economists (such as Ben Bernanke, Mark Gertler and Simon Gilchrist though the ideas go back, at least, to Irving Fisher, writing in the 1930s) to describe the process of credit contraction when asset prices fall. Much of the borrowing from banks and other financial institutions is backed by collateral, with both the amount of borrowing and credit terms such as interest rates or duration of loans depending on collateral values. When asset prices fall, collateral backing is reduced, tending to diminish credit availability and raise the effective interest rates at which finance can be obtained. In recent weeks the 'spreads' between corporate bond yields and safe government bond yields have risen sharply, signalling that investors have become more concerned about corporate default risks and that credit lines from banks have contracted with the fall in asset values and lower growth expectations. The U.S. Federal Reserve Senior Loan Officer Surveys show credit conditions tightening in the U.S. throughout this year. Spending and therefore growth falls with credit contraction, providing an important aspect of the 'feedbacks'. For evidence of the empirical link between widening spreads and subsequent slowing of growth, see Mark Gertler and Cara Lown in the Financial Instability issue in 1999 of the Oxford Review of Economic Policy.

U.S. consumers have a particularly high debt exposure, with household debt to income ratios at the highest levels for 70 years. Spending has recently exceeded income because asset values were so high and consumers were confident about future income growth. Many consumers have been financing part of their spending by increasing borrowing. Indeed, default rates on consumer credit have been running at historically high levels, even in these years of record growth and low unemployment. Lower growth and lower collateral values are bound to result in a substantial rise in default rates from current levels, making banks more wary about lending. Lower share prices of banks are likely both to signal the rising proportion of bad loans, and, by limiting the capital-raising opportunities of banks, restrict credit supplies further.

The process we are describing, nevertheless, suggests a fairly gradual slowdown in economic growth rather than 'falling off a cliff'. The U.S. third quarter GDP figures released in the last week are consistent with our description of the mechanism. They show business investment substantially down from the previous quarter but consumer spending growth up, though not at the highest rates of the last two years. The strength of consumer spending in the third quarter is explained by a mixture of factors. The broad share price indices were close to their peaks of the year at the end of August. House prices continued to rise, though at lower rates. Housing wealth supports consumer spending, partly because of the collateral it provides for borrowing. There are delays in the process: house prices tend to lag behind share prices and consumer spending related to housing wealth lags behind house prices. Recent consumer surveys taken in September and October show the sharp fall in confidence consistent with the coming slowdown in consumer spending.

Our analysis might appear to point to a downward spiral, such as seen in the gradual collapse of the 'bubble economy' in Japan in the early 1990s and the financial crisis in Thailand in 1997. However, we should make clear that the U.S. Federal Reserve is in a position to lower interest rates and halt the process. The U.S. economy is much more sensitive to interest rates than the Japanese (our research on Japanese consumer spending makes this clear). Thus, the U.S. will not suffer a decade long recession Japanese style.

Policy dilemmas for the central banks

The Federal Reserve faces difficult policy dilemmas. Oil-price induced inflation has been rising. The Fed is aware of the need for a slowdown in world demand to reduce the power of OPEC to sustain or raise oil prices still further. Another difficulty concerns the Dollar. If foreign investors reduce purchases of U.S. equities and companies sufficiently, the U.S. trade deficit could not be financed at the existing exchange rate, putting the currency under pressure. Furthermore, note that U.S. corporations, who have been borrowing in EURO's and switching to Dollars to finance activity outside the Eurozone, have made huge returns from this activity in the last two years as the EURO has fallen against the Dollar. If they believed that these returns were about to vanish or even reverse, so making borrowing in EURO's more expensive than in Dollars, they would have an incentive not to engage in this type of borrowing and may even want to lock in the currency gains made so far by selling Dollars for EURO's. Given the recent extremely negative sentiment towards the EURO, when the turnabout comes, there will be very sharp swings in the exchange rates. Many investors have been holding 'short' positions in the EURO, and as they they seek to correct these positions, they have to buy EURO's even as the EURO rises, causing it to rise further. The last week of October may well have seen the beginning of this process, with the EURO rising from just over 82 cents to over 87 cents in the week to November 3, and currently hovering at just over 86 cents. The Fed will be very reluctant to lower interest rates in a period of Dollar declines because of the risk that the decline will feed into higher commodity prices and higher import prices and so raise inflation. These risks would be much reduced if this also happened to be a period of falling oil prices but there is no guarantee of such a lucky coincidence. Signs from OPEC and Palestine promise little relief, as yet.

A decline in U.S. share prices has knock-on effects around the globe. Similar tendencies in terms of higher oil prices reducing growth are at work elsewhere. Indeed, growth in Western Europe, with the exception of the U.K. and Norway, is a little more oil price sensitive than the U.S. However, asset values have been more restrained in most of Europe than in the U.S., consumers and firms are generally less in debt and the EURO is certainly not overvalued. Furthermore, demand in Eurozone economies is less sensitive to lower share prices than demand in the U.S. (and the U.K.). It seems likely, therefore, that growth in the Eurozone will decline less than growth in the U.S. In Japan, where the banks have long been struggling to rebuild their asset base after so many bad loans, lower asset values make this task even more difficult. Growth prospects for Japan deteriorate sharply. Elsewhere, many emerging economies' stock-markets have already fallen even more sharply than those in the industrial countries as investors have sought refuge in safer assets.

The effects on U.K. house prices

It is possible that equity prices on the broad U.S. indices could decline by of the order of 20-25 percent from the peaks of earlier in the year (already more than achieved in the high-tech indices - the Nasdaq index is currently down over 40% from its high). However, in our view, the probability of this occurring by Christmas is low. The experience of the last few years is of an end-of-year stock-market rally and it seems likely that there are considerable amounts of liquidity ready to invest in shares. In the U.K. and Europe, there are also good reasons to be less pessimistic than for the U.S. Nevertheless, an early return to high values as occurred after the 1987 stock market crash seems unlikely given our explanation of why share prices have fallen and of the lags at work. A fall of, say, 20 percent in the value of U.K. households' equity holdings would reduce average U.K. house prices by 7-9 percent, assuming unchanged incomes and interest rates. On balance, we believe growth will fall more in the U.K. than in the Eurozone: U.K. growth is more sensitive to falls in asset values (see Maclennan, Muellbauer and Stephens). but somewhat less sensitive to high oil prices. However, Sterling is overvalued against the EURO and the U.K. is running a considerable trade deficit, though higher oil prices have recently helped. U.K. interest rates are likely to fall in response to falls in equity prices and increased fear of recession. However, with Sterling tending to follow the Dollar down, a decline in Sterling makes it harder for the Bank of England's Monetary Policy Committee (whose job it is to meet inflation targets), to cut rates. It is plausible that interest rate falls may only be modest therefore, though it is difficult to forecast interest rate movements at all accurately in these circumstances.

Thus, while we would not rule out an actual nominal fall in U.K. average house prices under such a scenario, the lags may be long enough to give time for interest rates to adjust to prevent such falls. However, at the upper end of the London market and in some of the more affluent areas in the South East, where share-ownership is high and significant proportions of owners work in the financial and the new hi-tech sector, it seems very likely, under this share price scenario, that house prices will fall in nominal terms. The tendency for rewards in recent years increasingly to be in the form of share options and profit sharing schemes is bound to be a factor in such declines, even if some end-of-year bonuses have already been guaranteed.

We conclude, that in the event of a decline in the value of equities owned by U.K. household of the order of 20 percent, which is not entirely improbable, U.K. house prices, particularly in London and the South East, are likely to fall next year.

It seems likely that commercial property prices will move in a similar direction and in similar locations.

A more optimistic scenario

Above, we have examined a relatively pessimistic scenario. However, a soft landing for the U.S. economy is certainly a possibility. It is possible that the reductions in demand around the world, induced by lower asset prices, tighter credit, greater uncertainty and higher oil prices, may operate quickly enough on oil prices so that substantial falls occur over the next few months. A reduction in Middle East tension would support such falls. The Federal Reserve and other central banks would then be less concerned about inflationary pressures and be more inclined to reduce interest rates, which in turn, would support asset prices. However, it is slower growth that is essential to achieving such a decline in oil prices and this would still suggest that equity prices cannot soon return to their peak levels. Indeed, one can argue that the Federal Reserve has in mind a target level of asset prices, substantially below the peaks of the last year, to bring about a rebalancing of global demand and supply for goods and services and so indirectly for oil. If the evidence for this rebalancing comes in soon enough, a sufficiently forward-looking Fed may start cutting rates even before oil prices begin to decline. It would thus reduce the risk of recession. The process described above would still play itself out, but in a milder form, with levels of loan defaults and corporate failures higher than in the recent past, but not high enough to raise serious worries about the profitability or indeed stability of major financial institutions. Our predictions for property markets would then need to be adjusted upwards somewhat, implying only a mild softening in prices from current levels.
Back