US Equity Deutsche Morgan Grenfell Research Portfolio Strategy Service

#34 Populist Capitalism And Other Wildly Bullish Themes

#35 The Economic Consequences Of The Peace

#36 Conference Call With Alan Blinder

#37 New Era Recession?

Topical Study #38 Fed's Market Model Finds Overvaluation

August 25, 1997 ❖

Dr. Edward Yardeni Chief Economist

E-Mail: [email protected] Web Site: http://www.yardeni.com/yardeni/ Three Ways To Correct Overvaluation. The folks at the have a new valuation model. It shows that stock prices are 15%-20% overvalued. I can’t disagree with this conclusion. The Dow did plunge 6.8% from a closing record high of 8259 on August 6 to 7694 on August 15, but then it rebounded just above 8000 the following three days. The Fed’s model suggests that the summertime correction may not be over, but I don’t expect much more than a 10% decline from the August 6 peak. I still expect to see the Dow hit 8500 by the end of the year, though it might take a bit longer to get there. I still expect 10,000 by the end of the decade.

History shows markets can stay overvalued and become even more overvalued for awhile. But eventually, overvaluation is corrected in three ways: 1) falling interest rates, 2) higher earnings expectations, and of course 3) falling stock prices—the old fashioned way to decrease values. Better-than-expected economic growth during the third quarter has already halted the recent decline in the 10-year . Another negative for bonds is the possibility that the German Bundesbank might raise -term interest rates to defend the deutschemark. However, lower-than-expected US inflation should stabilize the 10-year yield around 6.2%- 6.5% for now, before it declines to 5.8% closer to year-end.

Earnings of multinational companies may be negatively impacted by the strong dollar and weak overseas economies. Indeed, during the first week of August, both Coke and Gillette warned analysts that they should lower their growth assumptions for these two reasons. On the other hand, better-than-expected growth should boost profits of companies with revenues coming mostly from US sales. Profits in the technology sector should remain very strong especially as these companies benefit from increased spending by business and government to fix The Year 2000 Problem.

Greenspan’s Model Finds Some . , the chairman of the Federal Reserve Board, has developed a keen interest in the stock market. It was on December 5, 1996, that he worried out loud for the first time about “irrational exuberance” in the stock market. He did it again on February 25 of this year. He probably instructed his staff to devise a stock market valuation model to help him evaluate the extent of this irrational exuberance. Apparently, they did so and it was made public, though buried, in the Fed’s Monetary Policy Report to the Congress, which accompanied Mr. Greenspan’s July 22 Humphrey-Hawkins testimony. (I first wrote about it in the July 28 issue of my Weekly Economic Analysis.)

According to Mr. Greenspan’s new model, the S&P 500 should have been trading around 800 in July. Instead, it was over 900, or nearly 20% above fair value. This is the most overvalued the market has been since July 1991. In this prior episode, lower bond yields, rather than falling stock prices corrected the valuation problem. The Fed’s model shows that the stock market was 32% overvalued during September 1987, just prior to the October crash. Back then, the 10-year bond yield rose from 7.1% in January to 9.4% in September. By December 1987, the market was 7% undervalued mostly because prices crashed, but also because the bond yield fell and earnings expectations were moving higher (Exhibits 1, 2, 3, and 4). By the way, the market was fairly valued last December when Mr. Greenspan first mentioned irrational exuberance.

Page 2 / August 25, 1997 / Deutsche Morgan Grenfell Topical Study #38 Alan’s Model. As noted above, the model was buried in the July Monetary Policy Report. It was summed up in one paragraph and one chart on page 24 of a 25-page document.1 The paragraph is reprinted below. The Report’s chart shows a strong inverse correlation between the ratio of S&P 500 earnings—using 12-month-ahead consensus earnings estimates compiled by I/B/E/S International Inc.2—and the 10-year Treasury yield (Exhibit 5).

Of course, in the investment community, we tend to follow the Price/Earnings ratio rather than the Earnings/Price ratio. The ratio of the S&P 500 price index to expected earnings (P/E) is highly correlated with the reciprocal of the 10-year bond yield. The Fed’s model suggests that July’s fair-value P/E multiple should have been about 16 with the bond yield close to 6%. Instead, the actual P/E was over 19 (Exhibit 6). A 19 multiple would be fair value if the bond yield fell to 5.3%. The Fed’s analysis can be used to derive a fair-market price for the S&P 500 by dividing expected earnings by the bond yield (Exhibit 1). When the actual S&P 500 is greater (less) than the fair-value S&P 500 index, the market is overvalued (undervalued). In July, the market was almost 20% overvalued as noted above. This could be corrected without a price decline if the bond yield falls to 5.3% or earnings rise 20%–or a combination of a smaller decline in the yield and a smaller increase in earnings that add up to a 20% gain in the fair-value price.

Many tend to follow the P/E ratio using actual (four-quarter trailing) rather than expected earnings. There is a strong upward bias in expected earnings as measured by I/B/E/S. Collectively, analysts almost always overestimate year-ahead actual earnings. Yet the Fed’s model clearly supports expected rather than actual earnings for valuation purposes. There is always the risk that the market is even more overvalued than the Fed’s model suggests because expected earnings exceed both current and year-ahead realized earnings. The good news is that while the market is overvalued based on expected earnings, the gap between expected and actual earnings currently is relatively small (Exhibits 7, 8, 9, and 10).

Excerpt from Fed’s July Monetary Policy Report:

The run-up in stock prices in the spring was bolstered by unexpectedly strong corporate profits for the first quarter. Still, the ratio of prices in the S&P 500 to consensus estimates of earnings over the coming twelve months has risen further from levels that were already unusually high. Changes in this ratio have often been inversely related to changes in -term Treasury yields, but this year’s stock price gains were not matched by a significant net decline in interest rates. As a result, the yield on ten-year Treasury notes now exceeds the ratio of twelve-month-ahead earnings to prices by the largest amount since 1991, when earnings were depressed by the economic slowdown. One important factor behind the increase in stock prices this year appears to be a further rise in analysts’ reported expectations of earnings growth over the next three to five years. The average of these expectations has risen fairly steadily since early 1995 and currently stands at a level not seen since the steep recession of the early 1980s, when earnings were expected to bounce back from levels that were quite low.

1 http://www.bog.frb.fed.us/boarddocs/hh/9707Report.htm 2 http://www.ibes.com/

Deutsche Morgan Grenfell Topical Study #38 / August 25, 1997 / Page 3 - S&P 500 Valuation -

#1 1000 1000 FED'S S&P 500 VALUATION MODEL Jul

800 800 Fed's stock market S&P 500 Price Index Aug model shows some irrational exuberance. Fair-Value Price= 600 Expected Earnings divided by 600 10-Year US Treasury Bond Yield

400 400

200 200

0 0 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998

* I/B/E/S consensus estimates of earnings over the coming 12 months. All prices are monthly averages of daily close.

#2 50 50 FED'S S&P 500 VALUATION MODEL Aug

45 45

In the Fed's model, 40 40 the fair-value price Expected Earnings* is equal to expected 35 35 earnings divided by the 10-year bond 30 30 yield.

25 25

20 20

Aug 15 15

10 Fair-Value P/E=Reciprocal of 10 10-Year US Treasury Bond Yield 5 5 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998

* I/B/E/S consensus estimates of earnings over the coming 12 months. All prices are monthly averages of daily close.

Page 4 / August 25, 1997 / Deutsche Morgan Grenfell Topical Study #38 - S&P 500 Valuation -

#3 50 50 FED'S S&P 500 VALUATION MODEL* (percent) 40 40

30 30 Overvalued

20 20 Jul

10 10

0 0

-10 -10

Fed's S&P 500 -20 Undervalued -20 valuation model suggests stock -30 -30 prices were nearly 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 20% overvalued in * Ratio of S&P 500 index to I/B/E/S consensus estimates of earnings over the coming 12 months July. Overvaluation divided by the 10-year bond yield minus 100. can be corrected in three ways: 1) a drop in stock prices, #4 970 970 2) a drop in the Jul 890 S&P 500 INDEX 890 (1941-43=10, ratio scale) bond yield, and/or 3) 810 810 an increase in 730 730 650 650 expected earnings. 570 570

490 490

410 410

330 330

250 250

170 170

90 90 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998

Deutsche Morgan Grenfell Topical Study #38 / August 25, 1997 / Page 5 - S&P 500 Valuation -

#5 16 16 EQUITY VALUATION AND BOND YIELD

14 14 This chart appeared in the Fed's July S&P 500 Expected Monetary Policy 12 Earnings-Price Ratio* 12 Report to the Congress. Another, Ten-Year US Treasury more familiar way to Bond Yield 10 10 look at it follows:

8 8

Jul 6 6

Jul

4 4 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998

* Earnings-price ratio is based on the I/B/E/S consensus estimates of earnings over the coming 12 months. All prices are monthly averages of daily figures.

#6 20 20 EQUITY VALUATION AND BOND YIELD 19 Jul 19

18 18 There is a very good 17 17 correlation between 16 Aug 16 the P/E (based on 15 15 expected earnings) and the reciprocal of 14 14 the bond yield. The 13 13 yield suggests a P/E 12 12 of 16, well below the 11 11 market's 19 ratio. Is the market too rich, 10 10 or right to expect 9 9 lower yield soon? 8 Fair-Value P/E=Reciprocal of 8 10-Year US Treasury Bond Yield 7 7 S&P 500 Price to Expected Earnings Ratio* 6 6

5 5 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998

* I/B/E/S consensus estimates of earnings over the coming 12 months. All prices are monthly averages of daily close.

Page 6 / August 25, 1997 / Deutsche Morgan Grenfell Topical Study #38 - S&P 500 -

#7 30 30 S&P 500 P/E RATIOS

25 25

Using four-quarter trailing earnings 20 Q2 20 Jul

15 15

Using expected earnings* 10 10

Fed's valuation model suggests that 5 5 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 stock prices are determined by * I/B/E/S consensus estimates of earnings over the coming 12 months. All prices are monthly or quarterly averages of daily close. expected, not actual earnings. So P/E based on trailing #8 earnings is mostly 100 1000 S&P 500 INDEX & EARNINGS irrelevant to Jul valuation. 90 900

80 800 ______S&P 500 70 700 Price Index

60 600 Expected Earnings*

50 Aug 500 Trailing Earnings**

40 Q2 400

30 300

20 200

10 100

0 0 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998

* I/B/E/S consensus estimates of earnings over the coming 12 months. Prices are monthly averages of daily close. ** Four-quarter average.

Deutsche Morgan Grenfell Topical Study #38 / August 25, 1997 / Page 7 - S&P Earnings -

#9 55 55 S&P 500 EARNINGS: ACTUAL & EXPECTED

50 Aug Aug 50

There is a 45 Expected Earnings* 45 systematic upward (Pushed 12-months forward) Q2 bias in expected 40 Expected Earnings* 40 earnings as measured by 35 (4-quarter trailing) 35 I/B/E/S. Collectively, analysts almost always overestimate 30 30 actual earnings. Yet the Fed's model 25 25 clearly supports expected rather 20 20 than actual earnings for valuation 15 15 purposes. 10 10 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000

* I/B/E/S consensus estimates of earnings over the coming 12 months.

#10 180 180 S&P 500 EXPECTED EARNINGS* (as a percent of 4-quarter trailing earnings) 170 170

The good news is 160 160 that while the market is 150 150 overvalued based on expected 140 140 earnings, the gap between expected and actual earnings 130 130 is relatively small 120 120 now. Q2

110 110

100 100

90 90 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998

* I/B/E/S consensus estimates of earnings over the coming 12 months.

Page 8 / August 25, 1997 / Deutsche Morgan Grenfell Topical Study #38 Copyright (C) Deutsche Morgan Grenfell Inc. 1997. All rights reserved. The information contained herein has been obtained from sources believed to be reliable, but is not necessarily complete and its accuracy cannot be guaranteed. Any opinions expressed are subject to change without notice. Deutsche Morgan Grenfell Inc. and its affiliated companies and/or individuals may, from time to time, own, have positions in, or options on the securities discussed herein and may also perform financial advisory services, and/or have lending or other credit relationships with those companies. This material has been approved for distribution in the United Kingdom, to professional investors who fall within the exemptions contained within the UK Financial Services Act 1986 - Investment Advertisement Exemptions Order 1988, by Deutsche Bank AG London, 6 Bishopsgate, London EC2P 2AT. Member of the LSE and regulated by SFA, Tel: (171) 545-4900, Fax: (171) 545-4988. Orders placed by UK persons directly with Deutsche Morgan Grenfell Inc. will not be governed by all protection provisions of the UK Financial Services Act 1986. Additional Information Available on Request.