Why I'm Holding Dry Powder

Includes: BAC, BRK.A, BRK.B, GE, GS, SPY
by: Jeff Marston


Rothschild, Templeton and Buffett made fortunes by buying low and selling high.

In order to buy low, you need liquidity, aka "dry powder", during market drawdowns. This is a form of timing the market.

But the conventional wisdom says you should never try to time the market.

Let's look at the last two bear markets to see if market timing would have worked.

"Have faith in God, but keep your powder dry" -Oliver Cromwell

Buying stocks in a crisis is how mega fortunes are made. Baron Nathan Rothschild, who instructed investors to "buy when there is blood in the streets", waited until the English Stock Exchange had totally collapsed in 1815 before he started aggressively buying up financial assets at record low prices. During the Great Depression and while he was only in his early 20s, Sir John Templeton borrowed $10,000 and bought stock in all the firms whose equity traded on the New York Stock Exchange at a market price of less than $1. Templeton advised investors to "invest at the point of maximum pessimism". During the heart of the financial crisis in September and October of 2008, Warren Buffett's Berkshire Hathaway (BRK.A, BRK.B) bought around $10B worth of equity in high quality businesses such as Goldman Sachs (NYSE:GS), Bank of America (NYSE:BAC) and General Electric (NYSE:GE). Buffett knew how to be "greedy when others are fearful".

You've probably heard some of these colorful stories. On the other hand, you've probably also heard the more mundane market truism, repeated ad nauseam by professionals in the financial services industry, that you should never try to time the market. The conflicting advice presents a conundrum. If we try to "buy low and sell high" like Rothschild, Templeton and Buffett, isn't that a form of market timing? Perhaps the pros are right and we should stick to dollar cost averaging. Never try to hold cash in anticipation of a crash. After all, we must attempt to invest rationally, devoid of emotion. Or as Charlie Munger put it in 1995, "the right way to think is the way Zeckhauser plays bridge".

Richard Zeckhauser, aka "Mr. Probability"

The Method

With that in mind, what we want to do is look at previous bear markets and determine whether it would have been better, in rational, mathematical terms, to be fully invested in stocks, or whether it would have been better to go to cash early and then buy in after the crash. Now I appreciate that the process I used here is flawed in some important ways and I'll try to point out those flaws as we go. Flaws aside, I think the two case studies below are good enough to make some useful observations about the potential effectiveness of trying to time the market. As Keynes said, "it is better to be roughly right than precisely wrong".

Here's the first set of charts. Explanations will follow but you can see where we're about to go with this.

2008-2009 Bear Market

Start Period 4/1/2005 Peak 10/9/2007 Bottom 3/9/2009 Investor 3 Buys Stocks 4/1/2010 End Period 4/1/2015
S&P 500 level 1,164.43 1565.15 676.53 1,197.32 2,094.86

Total Return


$10,000 would become

Investor 1:

Worst possible timing




Investor 2:

Fully invested through whole period




Investor 3:

Ultra cautious market timing strategy




Investor 4:

Perfect market timing




Figure 1: 2008-2009 Bear Market

Figure 2: 2008-2009 Bear Market Simulated Results-CAGR

In the top column of figure 1, we have Investor 1. This hapless character tries to time the market but fails miserably. He stays in cash and achieves a 0% return from April 1, 2005 until October 9, 2007, the exact day that the S&P 500 (NYSEARCA:SPY) peaks at 1565.15. Then, he realizes his mistake at the worst possible time and bails back into a 100% cash position on March 9, 2009, the precise day that stocks bottom out at 676.53. By staying in cash after the bottom is in, he achieves a 0% return while stocks rebound sharply from March 9, 2009 until April 1, 2015, the end of the study period. The purpose of showing Investor 1's results is to try to find a "worst possible case" for market timing. Investor 1 is the bottom of our "bracket" as we attempt to create a closed system as opposed to a messy, open system.

Now before we go on, we should observe a couple important caveats. First, it is theoretically possible to do worse than Investor 1, and in fact much worse. In nature, there are no closed systems. For example, if Investor 1 had amplified his mistakes by doing things like trading on margin, using options and trading frequently, he would have been completely wiped out (and then some) in short order. However, for the purposes of these case studies, we'll assume that all of our theoretical investors are long term holders who don't use options or leverage and don't short stocks or even try to pick individual stocks. By stripping out these variables, we are limiting the applicability of the results to reality. I suppose that's the problem with theoretical models. In reality, variables are infinite but in order to get results that you can work with, you have to limit variables so that you can observe the effects that you want to measure without the noise of other factors.

Investor 2 is a purist. She has a 100% position in the S&P 500 for the duration of the 10-year period. She has an iron will and refuses to sell a single share even while the market is crashing. At the same time, she is exceptionally cautious and, perhaps fearing a double-dip recession, does not buy stocks at any time during the period. Another caveat is important concerning Investor 2. Her returns relative to the other three investors would actually be higher due to receiving more dividends. Dividend returns are probably the best reason to hold equities through a bear market. Her advantage due to dividends is most pronounced when compared to Investor 3, who stays out of the market the longest out of the four investors in the study.

Investor 3 attempts to time the market. However, unlike Investor 1, he is cautious about it and enjoys a modicum of success. He goes to a 100% cash position on April 1, 2005. Around this time, evidence was starting to emerge that the United States housing market was in a bubble. This wasn't a mainstream idea of course, but that's Investor 3 for you. He pays attention and is highly skittish about such things. Anticipating the crash early, he wants maximum liquidity for even the slightest possibility that a market crash may be afoot. He sells all his stocks way too early, and as a result, he misses out on almost three full years of stock price appreciation before the crash. Then, continuing with his cautious tendencies, he refuses to buy back in until the evidence of a recovery is overwhelming. He moves from a 100% cash position to a 100% stock position on April 1, 2010; over a year after the market bottom on March 9, 2009. Mind that this date is months after the Fed, Treasury, Congress and President (both of them, Bush and Obama) had demonstrated a commitment to support financial assets. At this point stocks had almost doubled from their March, 2009 lows and Buffett had publicly announced that he was aggressively buying stocks almost 18 months ago. Yes, the Occupy Wall Street protests were still going on around the country and everyone was complaining about the "shallow" recovery but overall the evidence was compelling that the worst was over. Even with his extreme caution, Investor 3's market timing enables him to realize almost exactly the same level of success as Investor 1 who stayed 100% invested through the entire period. Investor 3 achieves a 5.8% CAGR and a 75% total return.

Investor 4 is a superhero whose superpower is timing the market. He forms the top bracket of our closed system and is the type of market timer that definitely only exists in theory. He stays 100% invested in stocks until the exact day that the market peaks. He then sells and stays in cash until the precise day that the market bottoms at which time he moves to a 100% equity position. Therefore, Investor 4 actually has two stock holding periods. During the first one, prior to the market peak, he achieves a CAGR of 11.3%. For the second holding period, he buys back in right after the market bottoms and achieves an enviable CAGR of 25.4% from March 9, 2009 until April 1, 2015. His portfolio's CAGR over the entire 10-year period is 15.3%. Note that he achieved this result even though he was in cash and achieved a 0% return for almost two full years of the ten years in the study. Of course he could have done better if he used options and leverage but he's not the type of superhero who goes in for fancy stuff like that.

2008-2009 Bear Market Conclusions

What do you think of the results? I've got a few takeaways. First, I was impressed with how well a hold-at-all-costs strategy worked. Investor 2's portfolio enjoyed 6% compounded annual returns during a ten-year period that included the worst bear market in modern history- and that didn't even include dividends. That said, holding through that kind of drawdown would have taken some serious emotional discipline and I suspect that most passive investors like her probably sold when their portfolio started to turn bright red. Second, I was also surprised to learn that Investor 3's ultra conservative market timing strategy (5.8% CAGR) worked almost as well as Investor 2's passive hold-at-all-costs strategy (6.0% CAGR). Finally, it was surprising to see that whereas Investor 4 was of course pre-determined to win the contest, his real outperformance occurred after the crash. His returns during the last legs of the bull market that preceded the 2008-2009 bear market paled in comparison to his returns once the post-crash recovery started. This point is probably important to highlight since our current bull market is looking more than a bit long in the tooth.

2000-2002 Bear Market

Let's move onto the second case study- the 2000-2002 bear market that followed the dot-com bust. We'll keep the commentary and explanations short this time since we're using the same methodology as we saw in the 2008-2009 case study.

Start Period 12/1/1996 Peak 3/24/2000 Bottom 10/9/2002 Investor 3 Buys Stocks 10/1/2003 End Period 10/1/2007
S&P 500 level 743.25 1,552.87 776.76 1,117.21 1,539.66

Total Return


$10,000 would become

Investor 1:

Worst possible timing




Investor 2:

Fully invested through entire period




Investor 3:

Ultra cautious market timing strategy




Investor 4:

Perfect market timing




Figure 3: 2000-2002 Bear Market

Figure 4: 2000-2002 Bear Market Simulated Results-CAGR

Investor 1 attempts to time the market but fails. He buys at the absolute market peak and sells at the absolute bottom. He loses 50% of his investment.

Investor 2 stays fully invested in the S&P 500 through the entire period from December 1, 1996 to October 1, 2007. He enjoys 7.0% compounded annual growth in his portfolio. Note again that this result does not include dividends.

Investor 3 employs an ultra conservative market timing strategy. He hears Fed Chair Alan Greenspan pontificate on television about how "irrational exuberance" may be responsible for inflated equity valuations and he goes to cash in December, 1996. This is much too early in hindsight, and he misses the last three years of the bull market in the late 90s but he also misses the bust that started on March 24, 2000. The market bottoms on October 9, 2002 but he doesn't buy in until almost a full year later on October 1, 2003. He holds a 100% equity position until the end of the study period on October 1, 2007. His portfolio achieves a modest 3.0% compounded annual return.

Investor 4 is a perfect market timer. She cashes out at the top and buys in at the bottom. Her portfolio achieves a 14.0% compounded annual return. However, in this case, it is interesting to note that more than half of her total return is achieved in the last years of the tech boom in the late 90s. Her total return prior to the crash is 108.9% whereas her total return after the market bottoms is 98.2%. Like Investor 4, aggressive investors like Mark Cuban made real fortunes by staying in for the ride just long enough during the late 90s. More cautious investors like Warren Buffett decided to pass on the tech euphoria altogether but still managed to avoid the crash that followed.

2000-2002 Bear Market Conclusions

Let's make some conclusions about our second case study. Once again, the hold-at-all-costs strategy outperformed the highly cautious market timing strategy but the difference was more pronounced during the 2000-2002 bear market (7.0% CAGR for Investor 2 vs. a 3.0% CAGR for Investor 3). In addition, the bull market following this bear wasn't as long or pronounced and investors who went to cash too soon missed some big gains in the speculative euphoria of the late 90s.

Overall Conclusions

And if we put the two studies together, what can we discover? First, for patient investors who have the discipline to endure a bear market, a traditional buy-and-hold strategy will probably offer solid positive returns over the long run but with a few agonizing years mixed in. This is common knowledge and it should come as no surprise. Second, it appears that market timing isn't nearly as dangerous as many so-called financial professionals make it out to be. Even an extremely cautious timing strategy achieved a positive return over the study period in both cases. Further, in the 2008-2009 study, an extremely cautious market timing strategy was almost as effective as a buy-and-hold strategy. On the other hand, the only strategy that produced truly exceptional returns was an aggressive market timing strategy. For investors who are willing to hold a significant cash position in anticipation of a crash, this strategy has the potential to achieve returns that far exceed the results attainable via a traditional buy-and-hold strategy. Further, as we saw in both case studies, the only time when significant downside risk is present for an aggressive market timer is when he is fully invested in equities. In both case studies, Investor 1, (the horrible market timer,) only lost money when he bought at the top of the market and then stayed fully invested as the market crashed. This risk can be avoided by staying alert for signs that the market is vulnerable and holding large cash positions during these periods. By doing this, the market timer may miss out on some gains and may miss the last legs of a bull market. This may reduce his return by perhaps as many as a few percentage points. On the other hand, we're not talking about losing money here. We're talking about missing out on what are likely to be speculative, risky gains. In other words, the cautious market timer may give up a few percentage points of return to the buy-and-hold investor if he is too cautious. In exchange, the market timer gains the very real chance of achieving truly exceptional portfolio returns.

These are the conclusions I draw from looking at these numbers. For these reasons, and because I think there are signs that the market may be vulnerable to a crash sometime in the next 3 years or so, my largest position is cash. I look forward to hearing your thoughts. How much dry powder do you think is appropriate given today's market conditions?

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.