What Happens To Risk And Return At All-Time Highs? Part 2

| About: SPDR S&P (SPY)

Summary

Investing at ATHs is better on risk AND return than you might expect.

Return is improved because the strategy wins more often. ATHs occur during bull markets.

Risk is improved because the strategy loses less. Crashes tend not to be close to ATHs.

Lots of charts!

In an earlier article, I wrote that now, at all-time highs (ATHs) in the stock market, was a good time to invest. All-time highs provide lower-than-normal expected returns in the very short term (1-4 weeks), but higher-than-normal expected returns over somewhat longer periods (6-12 months):

And all-time highs have lower-than-normal risk, by several measures, on any time horizon up to a year:

A lot of my readers were incredulous. How is that possible? What happened to buying low and selling high?

I can understand the incredulity because, in the article, I posted the results of my investigations, but I didn't explain why they were true. And it is fair to ask for an explanation for counterintuitive assertions of any kind, but especially in the investing world, where money is on the line and dubious statistics are everywhere. So that is this article's task: to explain why, at ATHs, the risk is on average lower and the return is on average higher, than the usual risk and return in the market.

Breaking down the results

Many traders analyze their strategies with the help of the following equation:

Average_return = Win% * Average_gain + (1-win%) * Average_loss

That is, the returns to a strategy are a function of (A) how often you have a gain versus a loss in the strategy (the "win%"), (B) how much you make when you have a gain, and (C) how much you lose when you have a loss. You can improve on a strategy by either boosting your win rate, boosting your average gain when you win, and/or trimming your average loss. Any one of these things is easy to do, but it is hard to do two at the same time, never mind all three.

Trading strategies tend to lie on a spectrum. At the conservative end, you tend to get many small wins and the rare large loss. (Think investment-grade bonds or utility stocks.) At the speculative end, you tend to get a few large wins and many small losses. (Think out of the money options, penny stocks, and lottery tickets.)

Investing in the stock market on an average day, and holding for a period of up to a year, lies somewhere in the middle. Surprisingly, perhaps, investing at an ATH and holding for the same time period moves the needle toward the conservative, bond-like end of the spectrum. It increases the win rate and trims the average loss, at the expense of a cut to the average gain. (Note: for most of this article, I'll use a 6-month holding period, so as not to bury my readers under mountains of statistics.)

Investing and holding for 6 months

All days

ATH days

Difference

Average Return

4.05%

4.56%

0.51%

Win %

68.4%

75.8%

7.3%

Average Gain

9.82%

8.35%

-1.47%

Average Loss

-8.46%

-7.29%

1.17%

Click to enlarge

It is the improved win% that is responsible for the higher average returns, and the improved average loss that is responsible for the lower average risk, of ATH investing. So we can break down the question - How is it possible that investing at an ATH is better on risk AND return than investing on an average day? - into two questions.

1) Why does ATH investing have an improved win% over investing in the stock market on a regular day?

2) Why does ATH investing have a smaller average loss than investing in the stock market on a regular day?

Boosting the win percentage

There is a simple reason why ATH investing has a higher win% than investing on an ordinary day. Most ATHs occur during bull markets. (Think about it; of course they do.) And bull markets tend to go on for … a while. Thus most ATHs occur in a circumstance in which the market is headed up for the next "while." And so the win% on ATH investing is relatively high.

Click to enlarge

In the above chart, the thick blue line (left axis) is the S&P 500 index since from 1971 to June 30 of this year (so without the recent ATHs). The thin red triangles ascending from the floor (left axis again, but meaning days rather than index points) mark the number of days since a new ATH. The three big triangles correspond to long bear markets and recoveries starting in 1973, 2000, and 2007. The many small triangles in the 1980s, 1990s, and since 2013 show ATHs achieved in relatively rapid succession during the bull markets of those eras.

The green spikes along the x-axis (right axis) show the results of investing for 6-month periods beginning at each ATH. They tend to cluster, one ATH right after another, and these streaks of ATHs tend to start off with positive returns and then, as their bull ends, wind up with negative returns. The overall effect, however, is to give ATH investing a better-than-average win%.

Trimming the average loss

Investing at an ATH, for a limited period, sometimes loses, but when it does lose, it loses less than the average loss over the same investment period. The reason is that big market drops usually start later than, and take longer than, the limited period after the ATH.

We can illustrate this by looking at what happens to an investment portfolio that is in cash most of the time but invests in the market for a short period after an ATH. We'll call this an ATH_IN portfolio. An ATH_IN portfolio with an investment window of 6 months we'll call ATH_IN_126 (126 trading days in a six-month period). This is not a market timing method; it is not intended to show how to get outsized returns in the market. It's intended to show, rather, that when you invest in the window after an ATH, you rarely lose money, and then not a lot.

Click to enlarge

Here's the S&P 500 chart again (blue line, left axis, log scale), with the days since ATH on it (red triangles, right axis this time). The orange line shows the results of our portfolio. Most of the time when the market drops, the portfolio is out of the market for most of the drop. For example, during the large bear markets, the portfolio simply goes flat; it's out of the market. The counterexample is the 1987 crash; it happened soon enough, and sharply enough, after an ATH that our portfolio sustains the full drawdown. But 1987 is the exception.

OK, let's look at return anyway

From the chart above, the ATH_IN_126 portfolio clearly underperforms buying and holding. But that's not quite a fair comparison, since most of the time, the ATH_IN_126 portfolio isn't invested at all. For the time it's in the market, it actually does very well - better than the market on average.

To see this, I have to introduce the Compound Daily Growth Rate, or CDGR. This is just like CAGR, except for days rather than years. It's the amount by which the portfolio would compound daily if we measure from start to finish.

Since the ATH_IN_126 portfolio is in the market only 45% of the time, it actually compounds considerably faster than the buy-and-hold portfolio. So do several other ATH_IN portfolios:

Daily Returns on ATH_IN Portfolios

Time Period

Days In

%Days In

Total Rtn

CDGR (bps)

5 (1 week)

1640

14%

303%

8.50

21 (1 month)

2969

26%

276%

4.46

63 (1 quarter)

4374

38%

495%

4.08

126 (1 half)

5219

45%

656%

3.88

252 (1 year)

6290

55%

364%

2.44

Buy & Hold

11478

100%

2186%

2.73

Click to enlarge

The days after an ATH, on average, are not only less risky, but return more than average days in the market.

Exiting at ATHs is a terrible, horrible, no-good idea

Here's another way to think about it. If investing at an ATH were a bad idea - a risk-prone, negative-return idea - then we could improve returns over buy-and-hold investing by being out of the market for a period after reaching an ATH. To see whether this works, we need to consider a portfolio that is the reverse of the ATH_IN portfolio we just looked at. The ATH_OUT_126 portfolio is in the market except for the 6 months (126 trading days) following an ATH. If ATH investing is dangerous, this portfolio should outperform the index. Note that this is a market-timing suggestion, so looking at total return does matter.

Click to enlarge

As you can see, on a time scale since 1971, the portfolio does horribly, achieving about one-tenth of the index return. Nor does it do much to avoid risk. Notice the large drawdowns during severe bear markets. What the portfolio avoids is strong bull markets!

Nor is there anything special about a six-month period. Here are several ATH_OUT portfolios; all of them tremendously underperforming:

Click to enlarge

In short, staying out of the market after an ATH ("sell high!") is a great way to damage both the risk and return characteristics of your investments.

What the skeptics get right

What the skeptics get right is that the superior risk and return characteristics of the days after an ATH depend largely on the market regime. That is, most of the outperformance is due to the bull markets of the 1980s, 1990s, and since 2013. If we are in for a long choppy period, like in 1974, or standing on the verge of a precipice, like in 2000 or 2007, then investing in the wake of recent ATHs will underperform, obviously.

Having lived through the steep bulls of the 1990s and the mid-2000s, as well as the crashes that came after, I don't see many similarities to those high points today. There isn't the mania for stocks, nor euphoria in the markets, nor piles of speculative debt, that characterized both the dot-com bubble and the housing bubble.

So I don't see severe danger immediately ahead. I expect the typical slow grind of a mature bull market, with occasional pullbacks, for the next while. Defending that proposition, however, would take another article.

Nothing published here is to be construed as investment advice.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.