Someone who reads my blog emailed me this question:
“Cash vs. fully invested? How can a small-time investor take advantage of large market selloffs (2001, 2008) if you are fully invested all the time?
Sure you can sell a position to enter a better position, but you are likely selling at a loss. In a broad market sell-off, you ideally want some liquid capital to put to work at bargain basement prices on excellent companies.
So I guess the question is: If stocks sold off by 30% across the board in 2017, how would you take advantage of that if you are not holding any liquid assets to invest with?
As a small-time investor, we do not have a way to get our hands on insurance float to invest when stocks go on fire sale.
I realize holding lots of cash for long periods is not a great strategy either but what is the alternative if you want to have some dry powder for those few and far between events...
As a side note - I noticed Berkshire Hathaway is holding a record 85 billion in cash right now, so it seems like Warren is keeping some cash around for better opportunities than are available in the market today.”
This is a tough question. There are a few different answers. The theoretically correct answer would be to say that you should only hold cash when you cannot find a good enough opportunity. In other words, you should never have holding a certain percentage of your portfolio in cash as a goal. You should not have a strategy about whether or not to hold cash. You should just be selective about which stocks you buy. And so – naturally – when too few stocks are cheap enough to buy, you should end up holding cash.
That is the theoretically correct answer for a value investor. But I have some doubts about this one. Personally, I know the biggest mistakes I have made have come when I was holding cash. When the market is expensive, opportunities for value investors are scarce. This can push them to take risks. Forcing yourself to keep some of your money out of stocks – and in cash- until prices come down might help you avoid these mistakes. I have certainly thought about that possibility.
So, we could implement a rule. We could force ourselves to hold some money in cash purely because the market’s Shiller price-earnings (P/E) ratio or some other normalized valuation measure is too high. You could easily create a worksheet in Excel to tell you how much money you should have in cash. The easiest way to do this would be to look at historically high-priced years. So, you could look at 1999 or 2008 or 1965 or 1929. And you could ask yourself: what is the most money I would be willing to have invested in stocks during a year as expensive as that? Say you pick 1999 as the most expensive year you have ever seen. When the expensive peak hit would depend on exactly what earnings normalization method you were using. So, you might pick 2000 or 2001 as the most expensive year instead. But, in rough numbers – here is what you would probably see.
The dotcom peak would have a normalized P/E of something like 40, while the “normal” P/E over the entire history you were looking at (going back to 1929 or earlier) would probably be more like 15 times. That means that the difference between the average price level in a given year and the most expensive year on record would be about 25 times earnings. So, say you would not want to have any more than 50% of your money in stocks during a year like 1999. Well, you could take the difference between that overpriced year (40 times Shiller P/E) and the median P/E year (15 times Shiller P/E) and you would have 25 times earnings of overvaluation. Now, you should obviously be 100% invested whenever the market is fairly priced or underpriced. So, you would then just need to come up with a scheme where you divided the max amount (in percent) of cash you ever wanted to hold by the degree of overvaluation. In this example, you would just force yourself to put 2% of your portfolio into cash for every 1 point of P/E the market was overvalued.
Again, we are going to assume you want to be 50% in cash and 50% in stocks during a year as overvalued as the dotcom peak. A more aggressive market timer might want to be 0% in stocks and 100% in cash during a year as overvalued as that. But, I am assuming you are enough of a value investor – a stock picker for the long run – that you would want to have half your money in stocks even during a really overpriced market.
Okay. Now, take a look at those Shiller P/E numbers. So, assume a normal Shiller P/E is 15. And assume the Shiller P/E is now 27. That is 12 P/E points above normal. I said we should automatically put 2% of our portfolio in cash for every 1 Shiller P/E point above normal, so we would now be holding 24% cash under this scheme. That sounds about right. If someone told me they were keeping 24% of their portfolio in cash right now – I would say that sounds reasonable. If the Shiller P/E rose to 39, you would be holding 48% of your portfolio in cash. If the Shiller P/E ever rose to 51 – which I do not think it has ever hit – you would be holding 72% of your portfolio in cash. If it ever hit 63, you would have 96% of your portfolio in cash (63-15 = 48; 48 * 2 = 96).
I am not saying this is the best approach. But if you want a rule for how much cash to hold – this is the kind of rule I would create. I would determine what percent of your portfolio to hold in cash based on how many Shiller P/E points above normal we were at. If you think what I proposed is too aggressive a form of market timing – just cut it in half. Hold just 1% in cash for every Shiller P/E point. If you are a really aggressive market timer – more of a value trader than a value investor – then you can double my suggestion and make it 4% in cash for every 1 Shiller P/E point above normal. That means you would now have 48% of your portfolio in cash. If a value investor – a buy and hold type – emailed me saying they were keeping 48% of their portfolio in cash, I would say that sounds a lot like market timing. But, if a value trader – a market timer – said that they believed in market timing and wanted to practice it and that they had 48% of their portfolio in cash right now, I would say that sounds about right. If I was going to time the market, I too might put nearly half my portfolio in cash right now. The market is obviously too expensive right now. You should be holding much more cash than you normally would – if you are a market timer.
Are you a market timer?
Or are you just trying to reduce the severity of your mistakes? All I am trying to do is dampen the harmful shocks that hit my portfolio because of my own extra-high level of stupidity in stock selection during very expensive markets. So, for me, here is how I would implement holding cash.
I said you could make a rule like 1 extra point of Shiller P/E translates into 2% of cash in your portfolio. So, if we are at a 27 Shiller P/E and 15 is normal, you would hold 24% of your portfolio in cash (27-15 = 12; 12*2 = 24). That is fine if you want to sell things you already own. It works well for traders. It does not work as well for investors. I do not want to sell things I already own just to reduce market risk. I am looking to avoid stock picking errors. There is a better way for me to implement this scheme.
Say I like to hold five stocks in my portfolio and I like to keep them equally weighted when I first buy them. But, after that, I just let them run. This is not exactly how I buy things for my own account, but it is pretty close. I usually have about five stocks and a “normal” position for me at the time I buy it is about 20% of my portfolio. Now, I am aware that I make more mistakes when stocks are expensive, so I should add some sort of market risk modifier to my portfolio allocation. However, I do not want to sell a portion of all the stocks I already own. How should I – a buy and hold stock picker – implement this market risk adjusted cash level approach?
I can apply a market risk modifier to each purchase I make. Instead of holding 24% of my account in cash right now, I can just take 1 minus 24% equals 0.76. Now if I want to buy a new position, I do not make it a normal “full” position. I adjust my position size down to reflect the extra risk I am taking by selecting any stock in an overpriced market. Using my own account as an example, I like to buy 20% positions. But, with a Shiller P/E of about 27, I should make the position size 15% instead of 20% (27-15 = 12; 12 *2 = 24%; 1.00 – 0.24 = 0.76; 0.76* 0.2 = ~15%). This is the approach I would consider using. I would never use the portfolio wide cash level approach.
Why? Because I like to focus entirely on the buying of stocks. I do not like to think about asset allocation. I do not like to think about selling. I try to put all the risk management stuff into the stock selection process. This is what I suggest most people try to do. It avoids the trader mentality, it keeps you from wasting too much time thinking about the stock market, the economy, etc. and instead keeps you focused on picking stocks. I think this is important, but not because you cannot make money by making better predictions about the market, economy, etc. than other people can. It is hard to make better predictions about interest rates, the S&P 500, etc. than other people. But, it is also hard making better predictions about specific stocks than other people. What I am advocating here is focus. I think it is fine for people to invest based on the macroeconomy, or trade based on the macroeconomy, but I think those people should focus on either being stock pickers or asset allocators. And I think those people should focus on either being traders or investors.
Assuming that you are – like me – a stock picker and an investor, you want an approach that improves your stock selection process. You want to spend as much of your undivided attention – as much time as possible – on the buying of individual stocks. In fact, I want to spend every second I devote to thinking about investments to work that has to do with whether or not I pick a specific stock. I do not want to spend a single second thinking about selling, the overall market or cash levels in my overall portfolio.
To keep my focus on picking individual stocks, I would not set a cash level for the overall portfolio. I do know from my own experience – my own mistakes – that I should put less money in the stocks I pick when the overall market is expensive. So, something like this market risk modifier (1 minus 0.02 times the number of Shiller P/E points we are above normal) would make sense for me. I think it might make sense for you as well, but you would want to adjust it to fit your personality. The modifier can just as easily be 0.01 times the number of Shiller P/E points above normal. It can also just as easily be 0.04 times the number of Shiller P/E points above normal.
The only two hard and fast rules I would have are: 1) That you should set your cash levels based on the number of Shiller P/E points the market is currently above normal. And 2) that you set your cash levels at the time you buy a new position instead of forcing yourself to sell pieces of stocks you own and like just because the overall market went up.
Talk to Geoff about Whether Value Investors Should Hold Cash When the Market is Overpriced
Disclosure: No positions.
Start a free 7-day trial of Premium Membership to GuruFocus.
