I know I said in my last article that this one would be a thorough report on my portfolio performance but it isn't. I decided instead to write a word or two about the fact that some investors like to keep cash around for a market drop. I recently read an article by Aristofanis Papadatos in which he discussed this popular topic of being fully invested versus retaining some cash for when the market drops. Aristofanis made the argument that in the long run, it's more profitable to stay fully invested in stock than to withhold a percentage sitting in cash.
Not everyone agrees, of course. In the comments, one person stated that he is 70% invested in stock. The argument is that you will lose less when/if the market drops and that you can buy cheap stock with the cash you have at hand. I can see the reasoning behind both arguments, but I have yet to see the actual math since advocates for both sides seem to skew the numbers in their favor. One cannot assume that this is the year before a 50% market drop and likewise cannot assume there will be another 10 years of bull market.
So, let's stay objective and do the actual math. Is there a sweet spot for the perfect amount of cash to have available? Does it require the market to drop a certain percentage? How many years can you wait before a drop happens for it to be worth having the money at hand? I will try my very best to answer these questions.
There is, of course, another side of the argument for keeping cash and that is to not go bankrupt in case of a market crash. If you are retired or don't have a regular income which supports your regular expenses, I think it's wise to always keep a portion of your money in something other than stock, be it interest funds or cash in the mattress. This article is about the money you can invest safely, when the necessary securities have been taken out of the equation.
One could argue that cash can be good to have at hand to time sudden downturns in individual stocks. However, events only concerning one stock in particular are almost always based in changed fundamentals for the company, e.g. missed earnings or weak guidance. I will not go into this because such events must be analyzed individually and does not automatically present a golden buying opportunity. A broad market crash however concerns all companies, with or without changed fundamentals.
During the last 30 years, we've had three bad market crashes (1987, 2000 and 2008) which mean one crash every 10 years. I've heard the argument that the market moves more rapidly today and that we shouldn't include 1987 since the market is likely to crash more and more often. Fine, two crashes the last 18 years is one every 9 years. This doesn't change things too much. We might as well only look at 2008 and see that it was 9 years ago, meaning a crash will be coming later this year?
Statistically speaking, yes. History however, is not an accurate predictor of the future. If it were, we would have a huge sell-off right now in anticipation of the upcoming crash, thus creating a crash. No one can know if a crash is coming this fall or if we have another 10 years of bull market ahead of us. One can only do the math and then decide how much money one wants to bet with on a potential crash.
To make the calculations both manageable and relevant, I will look at the last 30 years. This means including three market crashes. There was a smaller downturn during a few months in 1990 (Iraq invades Kuwait) but since the market movements were contained within a year, it doesn't really affect the outcome of this article. Since there are an almost infinite number of portfolio constellations, I will only use the S&P 500 for the calculations.
As stated above, there is a chance of extremely bad timing when entering the stock market. Entering before "Black Monday" in 1987 would have lost you 30.75% of your money assuming you bought at the peak. Your portfolio value would have returned in 1991 not including dividends. If you'd entered in August 2000, you'd have lost 48.71% by September 2002 and not regained the value until December 2014, again not including dividends. And last, entering in September 2007 would have lost you 53.28% by February 2009, but you would be on even money in October 2013.
Looking at these numbers, it's not hard to see the argument for not being fully invested at all times but to save some cash for a 50% downturn (average maximum loss for the three mentioned was 44.25%). This requires the absolute worst timing possible though. If you, for example, had entered the market 2 years prior to the 2000 entry point described above, you'd only have had a value loss of 23% in 2002 and be on even money by December 2003.
Flawless market timing does not exist. It is just as improbable to buy at the absolute peak before a crash, as it is to buy at the absolute bottom. Being two months off in 2009 in either direction would have either gained you an initial 19.39% loss (being early) or an equally large lost opportunity in the other direction (being late). Therefore, using the tops and bottoms in the market to make the case for either strategy makes the whole argument irrelevant since such timing won't ever happen. The numbers are even less striking when you take the intra-year movements out of the equation and only look at the end-of-year results.
We must however account for dividends. I here assume that the fully invested investor continually reinvest his dividends each year while I assume that the investor holding cash will deposit said cash after a crash and then regain the cash reserve as soon as possible (waiting for the next crash). I will also when stating the value of the portfolio use the S&P 500 value from Dec 31st each year, not picking out the absolutes.
I've set up a few guidelines for the calculations.
- The initial amount invested is $50,000 in January 1987. This is then adjusted for inflation. E.g., the initial amount for entering the market in 2016 is $105,636.
- New money is deposited in January each year and is 10% of the starting amount from the previous year. E.g., if entering the market in 1987 with $50,000, the deposited amount in Jan 1988 is $5,000. This is also adjusted for inflation going forward meaning the deposit for 2016 is $10,563.60.
- Dividends are paid Dec 31st and are reinvested together with the new deposit in January.
- Cash reserves are rebalanced to the correct percentage every January except for the years following market crashes, where the reserve is set to zero.
This graph shows you how much money you would have had in your account on Dec 31st 2016 depending on which year you started investing in the S&P 500, following the rules stated above. I've actually used 4 years where the cash reserves are used. Obviously, 1988, 2002 and 2009 but also 2003 since the market continued to drop during 2002. Looking at the graph we immediately notice that there are no overlaps in the curves, meaning no matter how big a cash reserve is kept, the 100% invested investor would have the most money today.
This is due to the fact that there's now been enough time since the last market crash to recuperate the increased losses you'd have by not keeping cash outside the market. In fact, if we'd set the end date between 2010 and 2013, we would have seen that with an initial investment between 2000 and 2008, the accounts with cash reserves would have outperformed the fully invested account during a few years. The biggest difference would have been entering the market in 2008 (before the crash).
In this scenario, the 100% invested account wouldn't be ahead of the cash reserves until the end of 2013. This means 5 years after the biggest crash in the last 30 years, you'd still be ahead if you put 100% into stock. For the people entering in 2000, the 100% account would have lagged for a couple of years in 2004-2005 and then be ahead until 2009 and then ahead again in 2011. Below you see the exact same graph as above with an altered end date.
To add things up. If you have a longer investment horizon than 5 years (which I would recommend), you should always be 100% invested in stock. Keeping a cash reserve will inevitably drag down the performance of your portfolio. The counterargument would be to switch between being 100% invested and keeping a cash reserve depending on if the market is cheap or expensive. If there is someone who could pull this timing off, I want in on that crystal ball because experience tells us that most people trying to time the market buy expensive and sell cheap. An example from my 2 years of investing is 3M (NYSE:MMM).
When it was at $160, many deemed it expensive and "would never buy above $140" and "will wait for the drop." It is now above $200 and dividends keep trickling in. I believe this to be a prime example of the same philosophy we've researched in this article. Again, when looking at individual stock, there are a lot of factors to consider, but generally speaking, there's always, at any given time, at least one company worthy to invest in.
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.