Ok, having dispensed with the necessaries today, I can do a little bit of freestyling here.
There’s this creeping (or maybe “rapidly accelerating” is better) notion out there that we no longer need qualified people except in fields where their skills are unequivocally valuable.
This is perhaps most clearly evident in the election of Donald Trump and the appointment of folks like Betsy DeVos to ostensibly important positions.
The idea, generally speaking, is that outside of things like accounting and other disciplines where predicting stuff isn’t part of the job, so-called “experts” are just that – “so-called.”
This notion was around long before Trump made the term “so-called” great again in a tweet aimed at discrediting a federal judge who blocked the new President’s controversial immigration order.
Another way to conceptualize “so-called” is to use the term “pseudo”. That coveys the skepticism with which we should approach the work of practitioners of certain disciplines that don’t admit of an indisputable need for trained professionals (surgeons would be another example of folks to whom the terms “so-called” and/or “pseudo” most assuredly do not apply).
There’s perhaps no better punching bag when it comes to applying the derogatory “pseudo” label than economics.
Economists, like weathermen, are horrible at their jobs. The same is generally true of equity analysts. And of political scientists.
I’ve long been a proponent of extricating the social sciences from the grip of methodologies that more appropriately belong to the hard sciences. Indeed, one of the reasons I stopped short of a PhD in political science is that I grew exceedingly weary of peer reviewed journal articles whose authors sought to apply statistics to human behavior. I thought I was getting into the discipline to study political theory. You know, Hobbes, Rousseau, Locke, etc. That’s the way they pitch it in undergrad. Once you get to the Master’s level they tell you something different. “If you want to study theory, go to an Ivy League school,” I was once told.
Be that as it may, the notion that we should do away with educated, trained people in fields where precise measurement and prediction is impossible is patently absurd. If you want to watch a proponent of doing away with “pseudo-experts” squirm, ask him or her what we should do with meteorologists. Weathermen are wrong all the time. They are, after all, in the business of predicting. But they unquestionably are better at predicting hurricanes than someone who isn’t a weatherman. The existence of weathermen seems to prove that there isn’t a black and white distinction between “expert” and “pseudo-expert.” It’s a continuum, and figuring out where everyone falls on that continuum where one extreme is populated by accountants and brain surgeons and the other extreme is populated by folks like Deutsche Bank’s Joseph LaVorgna is well nigh impossible.
As investors, all of the above leads to one burning question: in the age of the internet where anyone can do their own research and disseminate it widely, do we still need sellside research given what we know about the penchant for sellsiders to be, well, offsides?
My contention is “yes.”
Anyone who tells you they don’t feel more confident reading a report that has the Morgan Stanley logo stamped at the top, an alphabet soup of letters after the names of the authors, and a long legal disclaimer at the bottom, than they do reading a blog post by a guy who professes to be a drug dealer from a television show is lying to you.
Now that doesn’t mean you wouldn’t rather read Heisenberg than you would sift through a boring ass analyst note. Nor does it mean that you might not occasionally get better information from Heisenberg than you get from analysts. But if your career was on the line, or if anything else of import to you was on the line, you’d cite the sellside note first. Why? Well because common sense tells you that even if the analysts turn out to be either wrong, or worse, completely corrupt and full of sh*t to boot, you can at least refer back to those letters (PhD, MBA, CFA, etc.) and the long legal disclaimer as proof that you consulted the “experts.”
But as true as it is, that’s obviously a sh*t argument for keeping sellsiders around. Let’s look at a better argument.
Beyond the “covering your ass by appealing to a possibly dubious authority” argument, there’s something else to be said for sellside research. Namely that it’s produced by people who have actually studied the subject matter and who, if they aren’t traders themselves, at least work with people who actually trade. Sorry, but that’s valuable. You can’t just write that off because they might be corrupt. Or they might be self-serving. Or, even sillier still, they might be wrong. Of course they might be corrupt. Of course their analysis is self-serving. And – for God’s sake – of course they might be wrong. That’s part of it. They’re human after all so they’re going to be wrong sometimes and they’re in the business of moving a sh*t load of money around, so of course they’re prone to corruption. I mean is that really news to anyone?
We need sellside research the same way we need historians. The same way we need psychologists. The same way we need psychiatrists. The same way we need political scientists. They’ve dedicated their lives to studying these subjects and whether you put any stock in their degrees or not, they’ve gone through the motions to get educated. Do you know what that means? It means that no matter what you think about their merit, they know more about their area of expertise than you do and that’s valuable.
Now if you, like one increasingly prominent (and increasingly dangerous) thinker I know, want to write everyone who isn’t an accountant, a brain surgeon, or a dentist off as a useless “pseudo-expert,” then be my guest. And while you’re at it, have fun trying to be your own psychologist, psychiatrist, political scientist, MBA, CFA, economist, and yes, weatherman.
With that, I bring you the counterpoint from a piece that appeared yesterday in the Financial Times:
Sellside analysts may be a dying breed. The Financial Times reported last week that the number of investment bank analysts providing economic forecasts, and stock and bond recommendations, has fallen by a tenth since 2012. There were only about 6,000 analysts working in these jobs at the world’s 12 largest investment banks in 2016, down from 6,600 four years previously.
Tighter regulation and falling profits have certainly had an impact, though the numbers suggest that reports of the death of sellside analysis have been greatly exaggerated. Yet even if the numbers do continue to decline, there would be little cause for mourning.
For a start, most analysts’ research is not very good. In 1973 Princeton University professor Burton Malkiel claimed in his book, A Random Walk Down Wall Street, that: “A blindfolded monkey throwing darts at a newspaper’s financial pages could select a portfolio that would do just as well as one carefully selected by experts.” Many of those pages have disappeared, but research suggesting random selection would be as accurate as analysts’ predictions continues to proliferate. In 2013, for example, Nerdwallet, a personal finance website, found that 49 per cent of analysts’ ratings on the top 30 US stocks during the year were incorrect.
There are many reasons analysts get it wrong. Like other people, they tend to believe that the past is a guide to the future. In 2004 an influential academic paper found that analysts from sellside firms generally recommended “glamour” stocks, that is those with positive momentum, high growth and high volume. Unsurprisingly, these also tended to be the most expensive stocks.
Analysts do not like to break away from the herd. Being an outlier and wrong is much more painful than being close to what everyone else is predicting and wrong. This discourages originality of thought and means convergence to the consensus is the norm. Nowhere is this more apparent than in the inaccuracy of analysts’ recommendations of whether to buy, hold or sell a stock.
Numerous studies have found that analysts routinely overhype the companies they cover, and that “buy” recommendations outnumber “sell” recommendations, even when the market is heading for a meltdown. An analysis in 2015 by Bespoke Investment Group of S&P 500 stock ratings found that, of the 12,122 ratings, just 6.67 per cent carried a sell label. The temptations to be bullish are clear. Access to company management gets more difficult with a sell recommendation, and the relationship is often of primary importance to an analyst’s employer, the bank. And, as a former analyst wrote in the FT, “fund managers do not want to talk to analysts about their hold recommendations”.