Maybe you should be in 100% cash

This post has nothing to do with asset prices, valuation, or timing the market as the title may have led you to believe. It has to do with investor psychology and behavior. Over the years I’ve wondered if certain types of people would be happier if they didn’t invest in anything but cash. Not ‘better off’ mind you just happier and still able to meet their financial goals – like a successful happy retirement. Then I said, “I have the data for that analysis”. Let’s take a look at the kind of people I’m talking about. You’ll probably see a bit of yourself in my description. Then let’s see what kind of retirement such a person could reasonably expect and some strategies to make it better.

I think we all know the type of person I’m alluding to. Most investors have some of these traits. Constantly worried about any kind of investment. Stocks – they’re always too expensive or so cheap it’s an indication of some forthcoming dire event and thus they must continue to go down. International stocks – same thing, even worse. Bonds – even the mighty US government is going default for sure. Any day now. Gold – sure, we gotta have a lot of that but I need to check the prices three times a day. And anytime prices go down it’s manipulation. Inflation – we’re constantly falling behind in standard of living. At the extreme, always worried about large inflation any day now. Yield – the need to reach for yield and check news every day that may affect the income stream. Price fluctuations of any significant amount are a sign to take action and seek refuge. Logging in to investment accounts way more than neccessary and checking account balances. Glued to financial news of any kind. Tweaking investments all the time always looking for the better bet. In general, constant unease about the future and definitely not able to sleep comfortably at night. I may be exaggerating a bit but I know quite a few people, young and old, that would fit a large part of this description. And I think a large part of these people can overcome these behavioral obstacles, especially by adapting an automatic investment process or system like the ones I discuss on this blog. But I also often wonder if some people would be better off just sticking with investing in 100% cash and never taking any investment risk. They would be much happier. Let’s take that as a given and see how much could such a person expect to reasonably withdraw from their cash portfolio in retirement (which also determined what size portfolio such a person would need to retire). Let’s find out.

Using the database I use to calculate SWRs (see here for an example), I replaced the US 10 year bond returns with the historical series for the US 3 month T Bill from 1929 to 2014 to represent cash returns. This is a pessimistic return series to use for cash returns but it’s the best series with that much history. Normally, even in environments with very low US T Bill rates an investor can get cash returns out to 1 year that are quite a bit higher. For example, even with today’s low T Bill rates of 0.14% or so you can get a 1 year CD at many banks yielding over 1%. Below are the historical SWRs (Safe Withdrawal Rates) for a few scenarios with a 100% cash portfolio.

The historical SWR for a 100% cash portfolio is 2.3% using a normal inflation adjusted spending model. That is quite a lot lower than the 4% from a 60/40 US stock bond portfolio but definitely do-able. And definitely a portfolio that would have allowed for many more restful nights. The rest of the table shows what the SWR would be with some tweaks to the spending model. The FCM  (floor-ceiling) model adjusts inflation adjusted spending down during bad return years. The spending adjustment column uses the historical fact that retirees’ spending increases less than inflation, between 1-2% less than inflation in fact. Using these better spending models increases the historical SWR to 3.34%. Doesn’t sound too bad now does it? Not too far off from the old 4% rule. But how realistic are these spending scenarios?

In my opinion and in my experience the above spending scenarios are easily achievable. Think about what people did before easily accessible and low cost investment options. Investing was something reserved for the wealthy or least very well to do. It’s only in modern times that investing is so widespread and accessible. How did your parents or grandparents plan and survive retirement? If they were like my grandparents they planned and survived retirement through a combination of saving a lot and not spending a lot. There was never any investment talk. Getting them to trust bank CDs took almost 10 years! Yet they made it and were quite happy along the way. Sure, they could have been ‘better off’ but they wouldn’t have been as happy. Now lets turn to a modern and more tangible example, me and my wife Nina.

From our base spending level in 2005, we spent 53% less in 2014. Yes, that involved a massive life change. By choice. You can read out it on Nina’s blog and even watch a little video about it. But it was for the better. Infinitely better for us. Oh, and that is in nominal terms. In real dollars, we spend 70% less (inflation has grown 2005 dollars by 20%) than we did in 2005. OK, that’s cheating a bit. At least in the ability to generalize from a very specific and personal choice. So, lets take our spending change since we started RV’ing in 2010. In 2014 we spent 10% less than we did in 2010. Inflation is up 8% since then. In real dollars that means we spend 18% less than we did in 2010. That’s over 3% a year. Obviously that can’t go on forever. And we’re pretty much at the bottom of the curve so to speak. Any further dramatic changes would require a reduction in quality of life which is not acceptable to us. Going forward our goal is to keep spending flat in nominal terms. Worst case to keep pace with inflation. I think that is pretty achievable. Our example just goes to show that controlling your spending so that it grows less than inflation is certainly achievable and not just data from some impersonal random study.

The other aspect of future spending is that most people have some type of retirement or pension income that begins in later years. This is mainly social security. So, in order to get a truly realistic picture of the future we need to forecast cashflows on a yearly basis. Then we can get a true picture of what SWRs would look like from a 100% cash portfolio. Kind of like I talked about in this post. Let’s consider a 65  year old couple just beginning retirement, delaying social security until 70, a median social security income of that covers about 40% of their expenses, and controlling their spending so that it grows at 1% less than inflation. Taking this cash flow model and applying it to the historical returns from a 100% cash portfolio gives us a worst case SWR of 4.13% for the 30 year retirement period starting in 1942. Not so harsh a retirement after all. Even in 100% cash. And definitely many more restful nights than an equity heavy portfolio.

In conclusion, sometimes taking an extreme position can be quite thought provoking and insightful. Admittedly, that is what I’ve done here. It was also a bit tongue in cheek. I’ve shown that even with a 100% cash portfolio a reasonable retirement can be had by focusing on the other side of the equation, spending, and using some more realistic retirement assumptions. People have been doing it for a long time. A lot longer than they have been investing in broadly diversified portfolios across world wide asset classes and markets. And maybe this thought experiment allows us to worry a bit less about our investments and have some more restful nights for just having thought through these alternative scenarios.