Is diversification better

Question

I have had a retirement account of one form or another for over 20 years. I lost a lot of money when the .com bubble burst and since then I have followed general rules about diversification. I have rebalanced yearly and kept my ratios in line with my age and profile. I recently reviewed my account, looking at the last 10 years (2007/09/01 - 2017-09/01).

What is disappointing is that if I would have put all of my money in my favorite fund for that time period, I would have almost 3 times as much money right now. For first 2+ years, I would have had less, but this fund has outperformed all of the other available funds and has a much longer track record (1960ish). I have a lot of questions that I've heard the scripted answers to dozens of times. I understand the theories, but I'm more interested about what has happened in the real world and what is likely to happen in the next 20 years and why.

Is diversification so widely recommended because most people are risk averse or because funds aren't generally reliable over time or is there real data proving it works best? Is a fund that has averaged double digit gains for over 60 years an anomaly? Is putting all of one's money in a stable mutual fund or an S&P 500 index fund really more risky than investing in several funds in different classes?

For a 40-something who isn't worried about yearly fluctuations, am I better off to risk not being able to retire because I lost too much investing in one fund or risk not being able to retire because I diversified and stagnated?

Edit:

For more context, my favorite fund is one that I selected more than 10 years ago as the one that would have the highest percentage of my contributions. It had the best and longest track record of any of the funds available and because it is small to mid cap growth, it fit with recommendations for my age and investment profile.

Answer 1

There are probably 3-4 questions here.

Diversification - A good index, a low cost S&P fund or ETF can serve you very well. If you add an extended market index or just go with "Total market", that might be it for your stock allocation. I've seen people with 5 funds, and it didn't take much analysis to see the overlap was so significant, that the extra 4 funds added little, and 2 of the 5 would have been it. If you diversify by buying more ETFs or funds, be sure to see what they contain.

If you can go back in time, buy Apple, Google, Amazon, etc, and don't sell them. Individual stocks are fun to pick, but unless you put in your homework, are tough to succeed at. You need to be right at the buy side, and again to know if, and when, to sell. I bought Apple, for example, long ago, pre-last few splits. But, using responsible a approach, I sold a bit each time it doubled. Has I kept it all through the splits, I'd have $1M+ instead of the current $200K or so of stock. Can you tell which companies now have that kind of potential for the future?

The S&P has been just about double digit over 60 years. The average managed fund will lag the S&P over time, many will be combined with other funds or just close. Even with huge survivor bias, managed funds can't beat the index over time, on average. Aside from a small portion of stocks I've picked, I'm happy to get S&P less .02% in my 401(k). In aggregate, people actually do far worse due to horrific timing and some odd thing, called emotions.

Answer 2

Diversification is the only real free lunch in finance (reduction in risk without any reduction in expected returns), so clearly every good answer to your question will be "yes." Diversification is good." Let's talk about many details your question solicits.

Many funds are already pretty diversified. If you buy a mutual fund, you are generally already getting a large portion of the gains from diversification. There is a very large difference between the unnecessary risk in your portfolio if you only hold a couple of stocks and if you hold a mutual fund. Should you be diversified across mutual funds as well? It depends on what your funds are. Many funds, such as target-date funds, are intended to be your sole investment. If you have funds covering every major asset class, then there may not be any additional benefit to buying other funds.

You probably could not have picked your "favorite fund" early on. As humans, we have cognitive biases that make us think we knew things early on that we did not. I'm sure at some point at the very beginning you had a positive feeling toward that fund. Today you regret not acting on it and putting all your money there. But the number of such feelings is very large and if you acted on all those, you would do a lot of crazy and harmful things. You didn't know early on which fund would do well. You could just as well have had a good feeling about a fund that subsequently did much worse than your diversified portfolio did.

The advice you have had about your portfolio probably isn't based on sound finance theory. You say you have always kept your investments in line with your age. This implies that you believe the guidelines given you by your broker or financial advisor are based in finance theory. Generally speaking, they are not. They are rules of thumb that seemed good to someone but are not rigorously proven either in theory or empirics. For example the notion that you should slowly shift your investments from speculative to conservative as you age is not based on sound finance theory. It just seems good to the people who give advice on such things. Nothing particularly wrong with it, I guess, but it's not remotely on par with the general concept of being well-diversified. The latter is extremely well established and verified, both in theory and in practice. Don't confuse the concept of diversification with the specific advice you have received from your advisor.

A fund averaging very good returns is not an anomaly--at least going forward it will not be. There are many thousand funds and a large distribution in their historical performance. Just by random chance, some funds will have a truly outstanding track record. Perhaps the manager really was skilled. However, very careful empirical testing has shown the following: (1) You, me, and people whose profession it is to select outperforming mutual funds are unable to reliably detect which ones will outperform, except in hindsight (2) A fund that has outperformed, even over a long horizon, is not more likely to outperform in the future.

No one is stopping you from putting all your money in that fund. Depending on its investment objective, you may even have decent diversification if you do so. However, please be aware that if you move your money based on historical outperformance, you will be acting on the same cognitive bias that makes gamblers believe they are on a "hot streak" and "can't lose." They can, and so can you.

======== Edit to answer a more specific line of questions ===========

One of your questions is whether it makes sense to buy a number of mutual funds as part of your diversification strategy. This is a slightly more subtle question and I will indicate where there is uncertainty in my answer.

Diversifying across asset classes. Most of the gains from diversification are available in a single fund. There is a lot of idiosyncratic risk in one or two stocks and much less in a collection of hundreds of stocks, which is what any mutual fund will hold. Still,you will probably want at least a couple of funds in your portfolio. I will list them from most important to least and I will assume the bulk of your portfolio is in a total US equity fund (or S&P500-style fund) so that you are almost completely diversified already.

1. Risky Bonds. These are corporate, municipal, sovereign debt, and long-term treasury debt funds. There is almost certainly a good deal to be gained by having a portion of your portfolio in bonds, and normally a total market fund will not include bond exposure. Bonds fund returns are closely related to interest rate and inflation changes. They are also exposed to some market risk but it's more efficient to get that from equity. The bond market is very large, so if you did market weights you would have more in bonds than in equity. Normally people do not do this, though. Instead you can get the exposure to interest rates by holding a lesser amount in longer-term bonds, rather than more in shorter-term bonds. I don't believe in shifting your weights toward nor away from this type of bond (as opposed to equity) as you age so if you are getting that advice, know that it is not well-founded in theory. Whatever your relative weight in risky bonds when you are young is should also be your weight when you are older.

2. International. There are probably some gains from having some exposure to international markets, although these have decreased over time as economies have become more integrated. If we followed market weights, you would actually put half your equity weight in an international fund. Because international funds are taxed differently (gains are always taxed at the short-term capital gains rate) and because they have higher management fees, most people make only a small investment to international funds, if any at all.

3. Emerging markets International funds often ignore emerging markets in order to maintain liquidity and low fees. You can get some exposure to these markets through emerging markets funds. However, the value of public equity in emerging markets is small when compared with that of developed markets, so according to finance theory, your investment in them should be small as well. That's a theoretical, not an empirical result. Emerging market funds charge high fees as well, so this one is kind of up to your taste. I can't say whether it will work out in the future.

4. Real estate. You may want to get exposure to real estate by buying a real-estate fund (REIT). Though, if you own a house you are already exposed to the real estate market, perhaps more than you want to be. REITs often invest in commercial real estate, which is a little different from the residential market.

5. Small Cap. Although total market funds invest in all capitalization levels, the market is so skewed toward large firms that many total market funds don't have any significant small cap exposure. It's common for individuals to hold a small cap fund to compensate for this, but it's not actually required by investment theory. In principle, the most diversified portfolio should be market-cap weighted, so small cap should have negligible weight in your portfolio. Many people hold small cap because historically it has outperformed large cap firms of equal risk, but this trend is uncertain. Many researchers feel that the small cap "premium" may have been a short-term artifact in the data. Given these facts and the fact that small-cap funds charge higher fees, it may make sense to pass on this asset class. Depends on your opinion and beliefs.

6. Value (or Growth) Funds. Half the market can be classed as "value", while the other half is "growth." Your total market fund should have equal representation in both so there is no diversification reason to buy a special value or growth fund. Historically, value funds have outperformed over long horizons and many researchers think this will continue, but it's not exactly mandated by the theory. If you choose to skew your portfolio by buying one of these, it should be a value fund.

7. Sector funds. There is, in general, no diversification reason to buy funds that invest in a particular sector. If you are trying to hedge your income (like trying to avoid investing in the tech sector because you work in that sector) or your costs (buying energy because you buy use a disproportionate amount of energy) I could imagine you buying one of these funds.

8. Risk-free bonds. Funds specializing in short-term treasuries or short-term high-quality bonds of other types are basically a substitute for a savings account, CD, money market fund, or other cash equivalent. Use as appropriate but there is little diversification here per se.

In short, there is some value in diversifying across asset classes, and it is open to opinion how much you should do. Less well-justified is diversifying across managers within the same asset class. There's very little if any advantage to doing that.

Answer 3

Diversification tends to protect you from big losses. But it also tends to "protect" you from big gains.

In any industry, some companies provide good products and services and prosper while others have problems and fail. (Or maybe the winners are just lucky or they paid off the right politicians, whatever, not the point here.) If you put all your money in one stock and they do well, you could make a bundle. But if you pick a loser, you could lose your entire investment. If you buy a little stock in each of many companies, then some will go up and some will go down, and your returns will be an average of how everyone in the industry is doing.

Suppose I offered to bet you a large sum of money that if I roll a die, it will come up 6. You might be reluctant to take that bet, because you can't predict what number will come up on one roll of a die. But suppose I offered to bet you a large sum of money that a die will come up 6, 100 times in a row. You might well take that bet, because the chance that it will turn up 6 time after time after time is very low. You reduce risk by spreading your bets.

Anyone who's bought stock has surely had times when he said, "Oh man! If only I'd bought X ten years ago I'd be a millionaire now!" But quite a few have also said, "If only I'd sold X ten years ago I wouldn't have lost all this money!"

I recently bought a stock a stock that within a few months rose to 10 times what I paid for it ... and then a few months later the company went bankrupt and the stock was worth nothing. I knew the company was on a roller coaster when I bought the stock, I was gambling that they'd pull through and I'd make money. I guessed wrong. Fortunately I gambled an amount that I was willing to lose.

Answer 4

As Warren Buffett says: diversification is a protection against ignorance. He says that diversification makes little sense if you know how to analyze and value businesses, because no singe person can properly understand the relevant details of a dozens of businesses. So, you should invest in just a few strong businesses, and not diversify your portfolio as that would amount to investing also in businesses that are inferior.

But even if we identify a few good businesses, wouldn't it then make sense to diversify a bit to lower the risk? Warren Buffett argues that a strong business are going to be be able to weather economic downturns and be able to deal with competition. He also argues then that there is actually less risk if your portfolio consists of a few strong businesses compared to a portfolio of 50 well known big businesses.