Click to Download as .pdf document

Last February we wrote about the possible outcomes for a market bottom. Would the technical pattern appear in the form of a traditional “V” shape or would it be a less dramatic turnaround such as a “U” (or “W”)? We then said that “it will take some time for the consumer to regain her footing and this will occur as consumer balance sheets stabilize. And given the rather high levels of household debt, the return to normal buying patterns will depend on mortgage refinancing and an improvement in the savings rate. This could be a longer process than most realize, and could lead us to consider an “L” bottom.” 

Looking back to the beginning of the second quarter, the S&P 500 had a sharp “V” type short-term recovery, but subsequently lost about half the gain. Potentially, it could still be in the downtrend that began last summer. Most likely it has entered a sideways market that could persist well into next year, much like what the Dow Jones Industrial Average experienced since April of 1999 when it was trading at 11250. Today, it’s at 9930! To visualize where the market will trade, it’s necessary to analyze the different economic sectors and forecast how sector profits are likely to advance. In the end, however, what the stock market does is not as important as what stocks do. As my friend and former associate Eugene Peroni Sr. used to say, “it’s not a stock market, it’s a market of stocks”.

Much of the bearish sentiment on Wall Street today is confined to the prospect for significantly lower consumer spending, which accounts for about two-thirds of Gross Domestic Product (GDP). There are many well-meaning analysts who are predicting a sharp drop in the market that would accompany such an economic event. This fear is being driven by a handy ratio called the “savings rate”, or savings as a percent of after-tax disposable income. This ratio, however, does not take into account the “wealth effect”. A study released in May by the Federal Reserve Board (1) shows that there is an inverse relationship between wealth and savings. (See Exhibit 1) Wealth is now such a dominating influence on the savings rate that the latter has become an unreliable barometer of the financial health of the consumer. The Fed’s research showed that the wealth effect drives consumption, albeit without a lag. Given today’s level of the wealth/income ratio, we suspect that the savings rate will be migrating upwards from the current SAVINGS RATE OF 2.5% to about 2.9%. (This is far below the 8.2% of the last recession that many other analysts are focusing on). 


Our model that measures the expected rate of change of real consumer spending suggests that the positive trend in force since the second quarter of 1997 will continue, just at a decelerating rate. Second quarter spending for durables was +5.9% vs. the second quarter of last year. Real spending on motor vehicles & parts was +5.5% and furniture (and household equipment) was +6.7%. Is this any way to behave when we’re having massive layoffs? Surely the growth rate will drop to a negative number in the second half . . . . or will it? I believe much of consumer spending is determined by demographics, the ratio of wealth-to-income, payroll growth and the unemployment rate. The latter is an important determinant of consumer behavior. The case in point is the “slowdown” of 1967 (Never an official recession!). The Federal Reserve had engineered an inverted yield curve (when short-term rates are higher than long-term rates) by the fourth quarter of 1966. The most recent inverted yield curve peaked in the fourth quarter of last year. Table 1 compares important consumer-related benchmarks between the 1966-67 slowdown and the current one.

This comparison is striking, and it is the low unemployment rate that makes these periods so similar. The key, however, to the current robust consumer spending cycle is the huge increase in real net worth that has occurred since 1997. (See Exhibit 2) There is no post-WWII business cycle that can compare to this one. It is this astounding increase in good fortune that has kept the consumer buying durable goods such as autos and furniture. Another positive for the consumer has been the mortgage-refinancing boom that began at the beginning of the year. This has enabled many households to unlock the equity value of their homes and increase their big ticket spending. And soon the check from Uncle Sam will arrive. But what about the forthcoming layoffs that we are hearing about?

The most volatile segment (other than construction) of the employed population is that of manufacturing, both durable and non-durable. Although today’s percent of the civilian population that is employed has risen from 57% to 64%, the percent represented by manufacturing workers has declined.

We often forget that we now live in a service economy, and that even when the industrial production is declining at rates not seen since the last recession, the influence on the employment rate is much less than one would think. Thus the effect on consumer spending of a manufacturing decline is not what it used to be. (Profits are another story!)

Another boost to consumer spending has been the continued strength of housing starts and real residential spending. With long-term interest rates peaking in January of 2000, housing began to pick up and did not get caught up in the jaws of the slowdown. This development supplemented the spending on consumer durables as well.

The negative impact of deteriorating net worth over the last several quarters has now flattened, and real payroll dollars growth should reach its low this half. Moreover the refinancing boom may not add incremental buying power unless interest rates fall significantly lower from here. Thus real consumer expenditures should bottom this quarter vs. last year’s third quarter. The unknown is whether the overspending of 1997 to 2000 robbed us of a more robust 2002-2004, and thus we might be facing an environment where the consumer simply stops buying. I think not. I believe a slower pace but sustainable pace of consumer spending is in the cards. Given the ongoing strength of consumer spending, there is little “recovery value” for most consumer categories at this point in time, however, there is no pending disaster either. 

So far we have accounted for almost 66.5% of GDP (before exports, imports, and inventory change.) No recession here. Government spending has been expanding as well, rising some 3% since the second quarter of last year. Can the 11.9% remaining be the cause of our profit woes? This is the area known as Non-Residential Fixed Investment. The trend of real nonresidential spending has been upwards since before 1960, increasing at a gradual pace until the early 1990’s when the productivity-enhancing technologies came on the scene. From 1992 to 2000, the contribution from this part of the economy almost doubled! (See Exhibit 3) Today, most nonresidential spending (73%) is made up of “equipment & software”. Equipment includes information processing equipment, software, industrial equipment, and transportation equipment. This area represents the heart of our manufacturing economy and industrial production. During 1990, technology spending was roughly 70% of non-technology spending; at the end of last year, it was 97%! 

The chart in Exhibit 4 depicts the trend of technology spending and its relationship to non-technology expenditures since 1982. Over the last two decades there has been an apparent link between spending on technology and labor productivity, so that the increasing dominance of technology is probably warranted. On the other hand, it appears that since 1999 there was considerable overspending on technology vs. other capital spending. In other words, the recovery in tech spending may be anything but robust over the next year. Furthermore, if one observes the relationship between capacity utilization and real nonresidential spending (Exhibit 5), it appears that the economy will need utilization rates above 78% before spending reaccelerates. This may be difficult to achieve because industrial capacity appears very overbuilt however relative to such measures as employees (See Exhibit 6). Unless there is a significant reduction in plant capacity, it may take several years before utilization rates exceed 78%. Technology sector utilization is now at 65.1%, a record low. There is little in the way of demand (new orders) to change the situation. At present the broadest cross section of technology reveals that through June, inventories have yet to turn negative (an historical precursor for recovery), while new orders are down almost 18% vs. a year ago (See Exhibit 7). In our view, trends for semiconductors are still poor, with inventories up 19% over a year ago, and a 3-month moving average of new orders heading down sequentially! As for the non-technology sector, industrial production just might be on the mend. The Conference Board’s Leading Economic Indicator series and industrial production have a good record of lead and lag relationships. In the 1990 recession, the LEI bottom occurred two months before non-tech industrial production bottomed. In March the LEI index appears to have bottomed, and three months later, non-tech industrial production appears to have bottomed (See Exhibit 8). It will take many more months to confirm this, but it is a positive sign.

When we put all this together, the economy appears to heading toward an “L” type recovery, with the bottom having already occurred, and with no significant “recovery” in sight until later this year. Most U.S. leading economic indicators have stabilized while Europe’s is still headed down. Without a more robust recovery, however, total corporate profits are unlikely to recover appreciably. This doesn’t mean that stocks are dead. Investors should not concentrate on the “market” but should instead be watching the “sectors”. Our analysis of the ten economic sectors of the S&P 500 suggests that earnings are likely to be somewhat divergent in 2001, depending on a sector’s sensitivity to capital spending and capacity utilization rates. Overall S&P 500 profit changes have historically been highly correlated with capacity utilization rates and crude materials prices at the wholesale level. Our current anticipation is that capacity utilization will recover only to about 80% at year-end 2002, and that crude material price increases will remain hard to come by unless the dollar weakens significantly. 
Investing in exchange traded funds such as the S&P 500 (SPY) and NASDAQ 100 (QQQ) was all the rage in 1997-2000. But now that the extreme valuations of technology stocks have been deflated, the thrill of broad-based index investing is gone! We recommend that clients and investors think not in terms of the NASDAQ 100, S&P 500, or even the Wilshire 5000, but instead concentrate on the most attractive economic, capitalization and style sectors. In our last update we wrote about the capitalization (large, medium, or small) and style sectors (value or growth). In this commentary, we want to focus on the economic sectors, highlighted in the table below.

Even though there are decent earnings recoveries for most groups in 2002, valuations for some sectors remain high and certainty low. Our performance forecasts for each sector are not only dependent on earnings, but also on valuations. In general, price/earning ratios vs. long-term growth rates are still excessive for some sectors, thus our year-end 2002 price change projections don’t necessarily correlate well with earnings changes. We look for the S&P 500 as a whole to post a gain in excess of cash equivalent returns over the next 12 months, but not by much. On the other hand, performance from sectors such as consumer staples, financials, health care, utilities and real estate investment trusts, should make pleasant reading. That is not to say that a small position in a less favorable economic sector is without merit. In the end, reasoned sector (stock) selection and reasonable diversification is the best strategy for controlling risk and generating good returns.