원문 보기
Another important question is will you bet on macro forecasts? Everybody makes macro forecasts and yet I don't believe that they are of any use. That's a mild exaggeration but not so great. Um the truth of the matter is that most of the time things continue as they are. So a forecast of extrapolation is correct most of the time and most people share that view.
Most people expect extrapolation. Extrapolation of the recent trends, the long-term trends is incorporated in security prices. So when the extrapolation turns out to be correct, it's not a profitable expectation because everybody else had it as well. it was in the price you paid. The forecasts that are very profitable are the ones that call for um uh radical deviation from trend, radical deviation from extrapolation.
These are very significant when you make them and they're right. The trouble is they're not right very often because most of the time things are extrapolated. And so we have forecasts of extrapolation which are usually correct but not profitable and forecasts of deviate uh deviation from trend which are potentially very profitable but not often correct.
I take that an analysis to mean macro forecasting doesn't work. And uh one of when we started Oak Tree 29 years ago this month uh we sat down and we wrote out the things we had been doing cuz we had the benefit of having worked together for 9 years already on average. And uh we have a very simple investment philosophy of six points which governs everything we do.
And number five says that our actions are not driven by macro forecasts. We are bottomup investors selecting individual individual companies not top down selecting markets and countries. So that's a different we think we can get an uh an advantage. By the way, where does outperforming come from? In short, it comes from having a nor a knowledge advantage.
And you have to know something better than most other people to to outperform. just trying is not going to help you if you don't know something they don't know. So um we think we can get a knowledge advantage in studying the micro companies, industries and securities not the macro countries markets and rates for example.
So you know we bar that from our approach. I one of my heroes is John Kenneth Galbrath who said we have two kinds of forecasters. The ones who don't know and the ones who don't know they don't know. And I've known for decades that I don't know. So I think we have an advantage in that. And then uh that kind of sounds like forecasting macro forecasting is unavail on availing but the truth is that it can be worse than neutral in its effect.
It can be damaging in its effect. And I quote the American humorist and novelist uh Mark Twain who said, "It ain't what you don't know that gets you into trouble. It's what you know for certain that just ain't true." No sentence that starts with I don't know but or I could be wrong but ever got anybody into trouble.
The ones that get you into trouble are the ones that say I'm sure that's such and such. Because if you believe something strongly and you bet on it heavily and you're wrong, that's how you get into big trouble. So, will you bet on macro forecasting? This, in my opinion, is a critical and and uh overarching decision.
I wrote a memo u a year and a half ago, I think it was October of 22, um entitled the illusion of knowledge. And it's a it's basically a rant against uh uh macro forecasting starting with a macro for a famous quote in each section. Um and uh I I think it's it's extremely important to decide will you or won't you start from a top- down macro forecast and most people do many people know they don't work.
They do it anyway. And one of the I quote somebody in there as saying that most people couldn't get out of bed in the morning if they admitted how uncertain the the future is. Uh and so I believe that macro forecasting uh in my view often takes the form of uh can I say organized religion. It gives something to believe in to dispel uh the great uncertainty uh present in the world they live in.
Um and uh you know I believe that uh very few people have the ability uh to do it in a superior fashion. John Kenneth Galbrath John Kenneth Galbrath for those of you who are not familiar with ancient history was an economist American economist uh died uh around 05 I think maybe a little after and uh at the age of about 98 and he was my one of my favorites.
He was he was uh a little on on the left side. Uh he was you know somewhere between free enterprise and socialism but um he was what you what we call a liberal in the days when when that word when it was okay to say that word. Um but he was he was very very smart and he was not a he was not famous as an economist but he was a he played a lot of roles in government and he was a diplomat.
Um but he wrote some very good books and one of them is called uh a short history of financial euphoria and uh uh I like thin books so and his is his is thin especially the ones he wrote in the last decade or two of his life were very thin so I enjoyed those but but the short history is very good and I recommend that to you um and one of the things he says is we have two classes of forecasters the ones who don't know and the ones who don't know they don't know now I don't believe in forecasts macro forecasts, people who forecast interest rates, performance of economies, performance of stock markets.
And I don't think that my efforts to be a superior investor and most other people's are aided by macro forecasts. Um, and uh, so am I saying that that the f forecaster is never right? No, I'm not saying that. The forecasters are often right. Last year, GDP grew 2%. Many forecasters forecast that GDP will grow this year at 2%.
That's called extrapolation. And usually in economics, extrapolation works. Usually the future looks like the recent past. So usually the people who forecast a continuation of the current are right. The only problem is they don't make any money because let's take let's take the economy. Most people forecast two something for this year.
A growth rate of two something is cooked into the prices of securities today. If the growth rate turns out to be two something, everybody who forecasted that will be right. But the security but security prices will not change much because that two something growth was anticipated and discounted a year or two ago.
So uh all those people who are right won't make any money. are forecast. So that's a cor that's a that's a correct. So most forecasts most extrapolation works all the time. Forecasts that are extrapolations work all the time but they don't make any money. Logically am I saying that forecasting that forecasts never make any money?
No. The forecast that make money are the forecasts of radical change. If if I predict if everybody's predicting 2.4% 4% growth for this year and if it turns out if I predict minus2 and it turns out to be minus2 or I predict six and it turns out to be six I'll make a lot of money. So forecasts which are not extrapolations forecasts which are radically different from the recent past are potentially very valuable if they're correct.
Of course they're not of any value if they're incorrect. And then if they're incorrect, they'll they'll cost you a lot of money. If if everybody else thinks it's going to be 24 and you predict six and it turns out at 24, you're probably going to have taken the wrong investments and lost a lot of money.
So uh uh deviant forecasts which turn out to be right are potentially very valuable, but it's very hard to make them. It's very hard to make them correctly. It's very hard to make them correctly consistently. And uh uh somebody at lunch mentioned an early memo I wrote called the value of forecasts. And uh in one of I there was the value of forecast and then there was value of forecasts too I think right and in one of those I reviewed the history recent history of the Wall Street Journal poll.
Every six months the Wall Street Journal uh publishes the results of a poll of economists and uh on you know GDP growth, CPI, value of a dollar, price of oil, whatever it might be, a bunch of phenomena they do it consistently and they they ask like 30 people consistently over time and uh so it shows basically that most of the time when people get it right, it's because they predicted extrapolation and nothing changed.
Once in a while something changes radically and invariably somebody predicted it. But the problem is if you look at that person's other forecasts over the years, you see that that person always made radical forecasts and never was right any other time. So of course if you if you're getting your information from a forecaster, the fact that he was right once doesn't tell you anything.
you you wouldn't the the the views of that forecaster would not be of any value to you unless he was right consistently and nobody's right consistently in making deviant forecasts. So uh so the bottom line for me is that forecasting is not valuable. Uh and uh that's something that my experience has told me.
So we ne we don't know what's going to happen and [clears throat] randomness will play a big role in what happens and randomness is by definition unpredictable.
Comments 0