"Most successful investments are where the odds are against you

Richard Zeckhauser, professor of political economy, Harvard Kennedy School

Published 11 years ago on Feb 02, 2013 11 minutes Read

champion Bridge player and the Frank Ramsey professor of political economy at Harvard University, Richard Zeckhauser calculates probabilities for everything that moves. So much so, that when we call to tell him that we might be about 10 minutes late for our 4 pm meeting at the Harvard Kennedy School campus in Boston due to an unexpected traffic jam caused by the construction work just across Charles River, he immediately begins calculating the probability of us calling him a second time to tell him that we would reach later than the anticipated 10 minutes. That’s the first thing that he tells us as we rush in to his office and apologise. “I put a low probability to it as you would have been too embarrassed to call about being delayed again.” Zeckhauser says calculating probabilities should become second nature and doing so would help improve our quality of business decisions immensely. As our meeting starts, he reminds us that his wife is scheduled to pick him up for a doctor’s appointment at 5 pm. As an afterthought he goes, “I think there is an 60% chance that while getting here she will get caught in the same traffic conditions as you guys, as she is driving over from this side of the river, too. That means we should have an additional 10 to 15 minutes.” Point made.

You wrote this much talked-about paper on investing in Unknowns and Unknowables (UU). What is the key takeaway in it? 

UU style of investing applies a lot to how corporations should invest in an uncertain world. Let me give you an example. There is an Indian restaurant around the corner owned by a famous chef from India — it’s doing okay. So the question is, should he have opened up this restaurant in the US where you think it is 50% likely that it will make money? If he makes money, he will make $100,000; but it is also 50% likely that he will lose money, in which case he would lose $200,000. That sounds like a pretty bad bet because you will make $100,000 half the times, that is, $50,000, and if you lose money half the time, you will lose $100,000. 

But look at it this way — if this restaurant made money, he could open up 20 restaurants across the US. By making the original investment where he is highly uncertain as to what he is going to do, he is getting an option of making not just $100,000, but 20 times $100,000. Actually, in this case, the uncertainty is his friend. 

Frequently, in a UU world, you can learn something from your original investment. Many UU opportunities have very strong unknown positives and very strong unknown negatives. What people should do is decide how big are the negatives and how big are the positives and how likely are they to come up. 

How do you go about assessing the likelihood of an event occurring?  

Let me give you an example. My wife went to see her doctor because she has elevated cholesterol. The doctor told her, “At your age, with your cholesterol level, if you take a statin, which is the standard medicine we prescribe, you will cut the probability of a heart attack in the next 10 years from 7% to 4%. But there are some dangers. The dangers are that one in 1,000 people has some muscle deterioration. Fortunately, we have a way of monitoring your blood. We can tell when that is happening to you. If you have this, you will know in a month and it is reversible. Another thing is that maybe one in 1,000 people has some degree of clouding up their mind. Once again, you and your relatives would probably recognise this. Then we would throw off the statin.” 

Many people will say, “With so many errors of omission and commission I really don’t want to do this, especially if it is going to cloud my mind, etc.” They don’t bother to think of what is the probability and is it reversible. The probability is overwhelmingly high that taking the medication could be helpful. But the reality is that people don’t like to take an active decision where it can harm them. 

The same thing is true for investments. So, let’s say an American corporation is contemplating whether to make an investment in India. Then, you hear about another American company that made an investment in India and then some major Indian conglomerate came along and influenced the chief minister of the state. The chief minister then retracted saying, “You can’t expand here” and it turned out to be a terrible story. When you get to know this story, you need to find out if that is one off or something that has happened a thousand times. By the way, this happens in the US, Canada, India and in every other country. I don’t know how often it happens in India but the wise thing would be to go and ask people knowledgeable about the country to tell you about all such cases and try to get answers. 

Take another example: many people come to me and say they would like me to invest in XYZ company, claiming it could be the next Microsoft. I have personally been offered 15 companies that could be the next Microsoft. My guess is that only a tenth of 1% of start-up companies could be Microsoft in the US — there has only been one Microsoft, after all. So the probability that I was getting the next Microsoft each time was one in 10,000. What are your odds of making money in such a case? Hence, you need to understand many of these concepts before placing your bets. 

When companies make acquisitions, they largely enter the UU world. Is there a way to think about acquisitions or any major corporate strategy moves to ensure a better success rate? 

In the US the general experience with acquisitions is that companies overpay. Their stocks go down and they don’t do well afterwards. In other words, acquisitions have unfavourable odds. Now what could be favourable odds here? It’s to say, I’m thinking of buying a company and if I offer to pay $50 million for it, I am 90% likely to get the company, but it’s only 40% likely to be worth what I paid for it. If I pay $40 million for it, I am only 60% likely to get the company, but I am 80% likely to have a profitable situation. 

So, there are less favourable odds on the first but more favourable odds on the second in terms of ultimately securing a profit. Decision theorists talk about using subjective probabilities all the time. You can infer those probabilities by looking at a variety of what you think are relevant cases.

How can companies prevent overpaying for acquisitions? 

There is this tremendous optimism bias built into acquisitions. Synergies in my experience are frequently overstated. If I were looking at a large merger, I would hire a team in my corporation to present arguments to the board as to why we should not do it. The idea is to have a countervailing team to poke holes in the logic. Organisations have this tremendous tendency to get behind the boss and do what he thinks should be done, but you have to get away from that and motivate people to bring to the table something contrary to what is being said. 

Otherwise, I would advocate going for small acquisitions at a low price that does not threaten your survival. There can be the occasional, highly strategic small acquisition to extend the marketing areas, where you offer 80% of what the target company thinks it is worth, but there aren’t many people who are interested in acquiring it. If it sells, fine; if not, that’s fine as well. Small acquisitions offer three advantages: one, you can get them at a more reasonable price; two, you get diversification; and three, you learn something from every company you acquire. By acquiring three little companies, you will learn more than you will learn from one big buyout. So, if it’s a small acquisition, I would worry about the price; if it’s a large acquisition, I would worry about how objective our analysis is. 

What do you mean by favourable odds in a world of UU where it’s impossible to calculate odds? Can you help us understand this with an example?

You have to think of a broader class of situations that are roughly equivalent to your situation. Say, I have a large business started by my grandfather and I’ve been running it; now the question is, how successful can I be in passing this on to my son now that the corporation has become much bigger? I, of course, think my son is terrific. But I start off by looking at situations where large companies have been passed on to relatives and then eventually realise that this is something that’s worked 30% of the time or it’s worked 5% of the time.

How do you make better decisions?

One part of decision-making is about how to place your priorities. Let me tell you what I said to a group of investment professionals recently. They were making investments and were being introduced to five fund managers. I said, “You have $50 million to invest and you have five potential managers; that does not mean you have to give $10 million to each of these managers. If you really think that manager A is much better, you should probably give him 25 and the others much smaller amounts.” Then, you improve your odds.  

Here’s another example out of what I see in everyday life. You get 50 e-mails during the day and you answer 30 of them. On the one that you answer the most, you take 3 minutes. In all the others, you take 45 seconds. You should take 25 minutes to answer the one that is important, but you don’t. Once that is pointed out to you, you will say that is really obvious. In other words, you should decide what is really important and make your choices accordingly. 

The other thing is about distinguishing between various probabilities. I think of making decisions the way I play tennis. I have taken many tennis lessons and my trainer always tells me the same three or four things. Keep your eye on the ball, get into position, swing your racquet back and swing the ball. I pay him $75 to tell me “keep your eye on the ball” and he tells me the same thing over and over again because the natural tendency when you are playing tennis is to take your eye off the ball. The natural tendency when you are thinking about probabilistic situations is to marginalise probabilities — treat 1%, 5%, 10% and 15% probabilities all as low probabilities. I think it is worth your while before you take a decision to figure out whether it is going to be 1%, 5%, 10% or 20%. And when it is worthwhile and when it is not. But most people don’t bother to do that. 

I am writing a paper today where we start off talking about President Obama’s assessment of the likelihood that Osama bin Laden was in the hideout where we found him to be. He had a variety of assessments and he eventually concluded well it was 50% likely that we were going to go get him. Now, there is nothing magical about 50%. It might be that it is perfectly worthwhile to go and raid that compound if the probability is only 30%. And maybe it is not worthwhile even if it is 70%. Think about that. But people feel that 50% is magical and they don’t like to do things where they don’t have 50% odds. I know that is not a good idea, so I am willing to make some bets where you say it is 20% likely to work but you get a big pay-off if it works, and only has a small cost if it does not. I will take that gamble. Most successful investments in new companies are where the odds are against you but, if you succeed, you will succeed in a big way. 

If it is about guessing probabilities, then, are we not playing too much on chance? 

Life is all about chance. There is no way I can avoid chance, but the idea is not to take a chance on your entire livelihood. I am a pretty serious bridge player, as are Warren Buffett and Bill Gates. Buffett says bridge is the best preparation for business. That’s because a hundred times in a bridge session you have to make a decision that depends upon chance. So you get comfortable doing it and that is the way life functions. 

Apart from playing bridge, how do you develop skills to make better decisions and think probabilistically? 

There are elementary books on decision analysis where they will tell you how you should be thinking about various situations. The best one I would recommend is a book called Thinking Fast and Slow by Daniel Kahneman that talks about all the problems people have making good decisions. 

I think that you can train yourself to make better decisions. At first, you may think it’s so complicated, but once you have made one or two decisions, you will get better. Daniel is more sceptical and, by the way, he is not a good decision maker. He is a brilliant scholar but he falls prey to all his anomalies! So it’s not just about knowing the concepts — you have to fight fiercely to avoid the mistakes. And then, play lots of games. You want to play engaging things in everyday life that resemble investing. 

How can companies use sidecars effectively, especially in the corporate world where there is adequate information efficiency and partnering with whoever has an advantage always comes at a price? Similarly, how can sidecars really work for investors? 

I would make two points. In the corporate world, there is barely adequate information efficiency. I don’t think of it as going into a supermarket and knowing what you are getting when you are getting a can of soup. But I agree, getting the advantage always comes at a price. But think about it what would you prefer — a car that breaks down more often but comes at a cheaper price, or one that is somewhat more expensive but will perform better? 

You are just riding in the sidecar because you know the route. For example, I don’t know much about real estate but am partners with this guy who I think understands real estate very well. He thinks I am very good at thinking about risk and uncertainty, borrowing strategies and the like. So it works very well. That is the ideal situation, where I give you something that doesn’t cost me much and you give me something that doesn’t cost you much.

The best sidecar investment that I can think of is Buffett. People really know how good an investor he is, so that gives him a lot of legitimacy. Not only can he go and buy into a company, he can get a more favourable deal than anybody in the world.