October 22, 2002
Time to Buy Equities?
Time to Buy Equities?
Jeremy Siegel believes that the bear market is over. Or, rather, he believes that the large equity premium return that stocks pay now more than outweighs any chance of a further large stock market decline. In his view, it is not a bad time to buy stocks--even though he believes that it is a worse-than-average time to buy stocks.
Dive Right in... the P/Es Are Fine: "You're starting from a much lower base," says Siegel. "I don't consider this market to be dirt-cheap, but it is a good, if not better-than-average, time to buy equities."
Siegel's full text
Posted by DeLong at October 22, 2002 12:55 PM
From the article: Siegel is just as bullish now as he ever was, insisting that investors can confidently expect to make an average of 5% a year after inflation over the next 20 to 30 years. The main reason? Today's buyers are getting in at prices more than 40% lower than in March, 2000. That matters because the five big market busts of 40% or more over the last century were followed by above-average annual real returns of 8.6% over the next five years. "You're starting from a much lower base," says Siegel. "I don't consider this market to be dirt-cheap, but it is a good, if not better-than-average, time to buy equities. So let me get this straight. Siegel is using 5 datapoints over the course of a century to make a "buy" case. Well, maybe, but anyone silly enough to buy into this reasoning deserves what they get.
Nice point about Jeremy Siegel stock push.
Whether right or wrong, what Siegel does not explain is why a 50% fall in the S&P over 31 months would leave the with a p/e over 20. How expensive was this market by January 2000? Were earnings so grossly inflated in reports at that time? Are earnings reports more realistic now? Why should p/e ratios be related to interest rates? After all, low interest rates may well indicate a slow growth outlook. This still appears a tricky time to invest, though we have sold bonds and bought the S&P index on dips. Does bond holding make any sense now?
Put differently, we might have asked, "Time to Sell Bonds and Property?"
Surely there is a time to make the switch, and we probably all have different views about what time is optimal. I have a hard time believing myself that there are further substantial gains to be had from holding bonds and property, and I admit I don't want to envision too vividly the awful marcoeconomic scenario that would prove me wrong.
After the 1990-91 recession, as the economy slowly reflated, it took a couple of years before the bond market eased, and when it happened with Fed rate hikes in 1994, the bond market got completely crushed. Maybe we will creep along for awhile this time, reading too many Jeremy Seigel type articles before finally reading the right one.
Well, I have sold bonds and bought the S&P.
Still, someone who held the Vanguard long term treasury fund these last 10 years would have done better than holding the S&P. Though bonds declined in price in 1994, the decline was "not" a crushing and lasted only about 12 months before the bond bull market resumed. I see no reason to hold treasuries now, but corporates of all qualities are better priced as is the Vanguard High Yield municipal portfolio.
Perhaps the most compelling point that Siegel makes is the favorable capital gains tax rate on stocks held for at least a year. I continue to be worried about pension fund accounting and options costs, but I am in stocks again.
October 14, 2002
Early in 2001, an independent study showed that while the annual return of the stock market itself averaged 16% per year during the 1984-2000 period, the return of the average mutual fund averaged 13%, about the differential one would expect, given that fund costs amounted to about 2 ˝% to 3% per year. But, because of the market timing and adverse selection issues I've just described, the annual return of the average mutual fund investor averaged just 5%. Today, the bear market has reduced that cumulative market return to about 11%, and the return of the average mutual fund to about 8%. That relationship suggests that the return of the average fund investor, during this excellent (from point to point!) period for stocks, was 0%(!). Nothing. It is not a record of which we should be proud.
What John Bogle addresses and Jeremy Siegel does not is who gets the gains on stocks. The need is for shareholders to truly benefit from investing, but thanks to financial intermediaries that are run for managers rather than for investors all too little stock market gains have been realized by us over the years.
Wall Street tells us of the paper gains from stocks, the real gains after Wall Street takes its share can be shockingly small.
Wall Street takes a brutal amount of what should be our returns and uses "ideal" returns to get us to invest with Merrill or Goldman or Prudential or Smith Barney and the like. Good grief.
How much do costs matter? Let's consider what costs mean to a long-term investor, and see what toll a 2 ˝% annual cost would take on a 10% stock market return over, say, thirty years. When compounded over thirty years, $1,000 earning the 10% stock market return would grow to $17,500. Pretty nice! That same $1,000 invested in the typical fund, earning 7 ˝% after costs, would grow to $8,750. Not too bad! But only half as much as the market—a dead-weight loss of $8,750 engendered solely by reason of the costs of financial intermediation. Put another way, the intermediaries put up 0% of the capital, took 0% of the risk, and garnered 50% of the return, the investor put up 100% of the capital, took 100% of the risk, and received 50% of the return. Put me squarely in the camp of those who don't think that's good enough.
That simple example reflects both the miracle of compounding returns and the tyranny of compounding costs. But the story gets worse. Intermediation costs are paid in current dollars, while the investor's final capital must be measured in constant dollars. Let's assume a future inflation rate of 2 ˝%. Result: Real annual return for the market, 7.5%; real return for the fund investor, 5%. The final purchasing power of each initial $1,000 falls to $8,750 in real terms in the market, and to only $4,300 in the fund. We're surely a long way from that $17,500 that appears on a table showing the magic of compounding 10% returns.
“Why should interest rates be related to P/E ratios?” If I remember my long ago finance/economic theory they relate directly. Although, since no one in the financial press seems to talk about that, I wonder what I’m missing/forgetting. A stock price is the theoretical present value of all future income flows to the investor, right? If we assume some correlation between the discount rate and long-term interest rates (reasonable, I think), plus an investor who’s not entirely short-term and irrational (questionable, granted), won’t lower rates mean higher present values? All other things being equal, of course, which is where economic concerns, etc. rear their nasty heads. The problem I have with the present value idea is correlating earnings to investor cash flows. How does the investor get his/her money out? There are limited ways: dividends (out of favor), company stock buy-backs, selling the stock to someone else (greater fool?)...
P/E Ratios are actually a shorthand for the Net Present Value (NPV) of future Net Income. (Actually discounted cash flows are the appropriate measure but it's the same principle so let's leave that issue alone for now). To arrive at the NPV of this future Net Income, you need to discount the returns from future rates. How do you arrive at this discount rate? The most widely-used approach is the WACC approach where the cost of equity is calculated using a number of measurements, with the rate of risk-free return being a prominent one. The conventional proxy for risk-free return are US government bond issues which, as we all know, are affected by the projected inflation rate. Hence, lower inflation will lead to a lower discount rate. A lower discount rate, because it is in the denominator, will lead to a higher NPV of future Net Income and higher stock prices.
Thanks for the note on p/eratios and interest rates. Yes, but… I have trouble buying the idea that stocks are cheap now merely because interest rates are so low. Low interest rates mean low future growth rates are expected. The p/e ratios now do not count options costs or pension fund earnings padding. Though I am in the S&P index again, I am feeling quite cautious. We may still have an “expensive” market. Besides, if John Bogle is correct, mutual funds have given poor returns to investors for many years and we should ask for a bigger risk premium. Investing in mutual funds has not been consumer friendly for some time.
Basically, technically speaking stocks are oversold. But fundamentally speaking they're over valued. It makes for an interesting dynamic. I would argue it's tradeable. But I wouldn't long-term invest in most sectors today.
It’s hard to know what low interest rates say about expected economic growth, except for the short term. Perhaps that problems are being addressed and we’ll get a turnaround? :-) So many things affect long-term expectations. I still think partially countering this muddled message is the direct effect of interest rates through the present value calculation. I found a sort-of proxy for that at: www.troweprice.com (they’re selling mutual funds, so beware.) In ‘Today’s Market: Still Overvalued, or an Unusual Buying Opportunity?’ they compared P/Es to the inverse of the 10-year Treasury bond.
As to earnings, if they’re dishonest it’s a rigged game and what rational person would play? Except few games are totally fair and you’ve got to play somewhere.
October 24, 2002
Looking Glass on Earnings Just Got Darker
By FLOYD NORRIS - NYT
Just how little is American industry making?
The Standard & Poor's Corporation will release today its calculations of "core earnings," as it defines them. That definition, which is subject to dispute, will make most companies look far worse than they have appeared.
The core earnings for the companies in the S.& P. 500 came to $18.48, S.& P. says, according to a copy of the announcement provided yesterday. That compares with operating earnings — the number Wall Street generally prefers to emphasize — of $41.58. And it compares with as-reported profits of $26.74.
Based on the value of the S.& P. of 989.52 at the end of the second quarter, the index was trading at a multiple of 54 times core earnings, a figure that has since declined to 47. By contrast, those using the operating income numbers see the multiple as 24 at the end of June, and 21 now.
Didn't I read yesterday that using the new S&P Core Earnings, the S&P 500 P/E is now 57?
Oops. I hadn't read the last entry when I posted.
A variety of models for long term real returns
on stocks--dividend growth models, simply taking
a cyclically-adjusted earnings yield, etc.--put
the long term real returns on the market at 4-5%
right now. If you're satisfied with that compared
to a 3% yield on long term inflation-indexed bonds
then you are a rare bird. Besides, the financial
services industry and Uncle Sam end up with the
vast bulk of the gains you might be lucky enough
to make. So pay off your mortgage, invest the
rest in TIPS, and get on with your life.