August 2002 Archives
What are stock analysts for?
Stock analysts, the CNBC superstars of the go-go 90s, have been back in the news of late. Henry Blodget, Merrill Lynch's star internet analyst, said one thing in internal emails and another in public reports, bringing down a $100 million fine on his firm. And Jack Grubman, Salomon Smith Barney's telecoms analyst, has recently quit his $20 million-a-year job, but still faces lawsuits alleging, most recently, that he upgraded his rating on AT&T only so that his firm could get a lucrative underwriting deal from the telecoms giant.
The general problem with such people, from the point of view of Eliott Spitzer or Gretchen Morgenson, is that they're duplicitous hyprocrites, who purport to be giving advice to investors on which stocks to buy when in fact they're just buttering up their firms' potential clients.
Spitzer, New York's ambitious attorney general, has received glowing press and is now better placed than ever for his forthcoming run for governor. Morgenson has already got her Pulitzer. It's easy to see why analyst-bashing is so popular right now: as greedy investors look at the ruined state of their stock portfolios, they love anybody who says that someone other than themselves is to blame.
Among people long familiar with Wall Street, however, the shock was not at the behaviour of the rogue analysts, but rather at the way in which the fearless press (not to mention the Pulitzer committee) was lapping it all up. After all, it was common knowledge during the boom that these analysts were earning tens of millions of dollars a year: where did people suppose that money was coming from?
I was set to thinking along these lines after reading a piece by former analyst Paul Kedrosky in the National Post. Stock analysts purport to be stock pickers, he says, but in fact they're not: if they were good at picking stocks, they'd be picking stocks rather than analysing them. Kedrosky claims that the purpose of stock analysts is to provide essentially a set of fresh eyes for institutional investors: maybe they'll see something the fund-manager missed, or come up with an interesting new angle. The actual rating - buy, sell, hold - is, on this view, irrelevant.
I think Kedrosky is hopelessly out of date, although things might be swinging back in that general direction. For one thing, ratings upgrades and downgrades do move stock prices, so they can't be quite as irrelevant as Kedrosky says. For another, Kedrosky seems stuck in the old world of financial markets, where companies just went ahead and did their thing, analysts analysed, and investors arrived at a collective decision for the value of the company. There were no feedback loops: a company's fundamentals were reflected in its stock price, but the stock price itself was not one of those fundamentals.
Most importantly, in the old days, stock analysts were modestly paid, mainly because they produced no revenue for their firm. The only way they could justify their existence at all was by invoking the honour of institutional investors, who apparently felt duty-bound to use Bank X's trading desk to buy or sell any stock which they were dealing in as a result of Bank X's research. Whether that was actually the case or not nobody knew, and in any event the marginal increase in brokerage commissions which the average analyst brought in was never enough to justify a seven-figure salary.
It was obvious, then, when analysts started becoming superstars and bringing home seven figures monthly as opposed to annually, that something significant had changed. And it was obvious, too, what that change was: the primary audience for analysts' research was no longer institutional investors, but rather the very companies they were covering. A hot analyst like Blodget or Grubman could create a buzz around a company, which would keep investors concentrating on the rising share price rather than asking awkward questions.
WorldCom was a prime example. It was generally considered a Bernie Ebbers story: iconoclastic CEO, through sheer force of personality, single-handedly shakes up the fusty US telecommunications industry and creates hundreds of billions of dollars of value in the process. But in fact it was just as much a Jack Grubman story: every time Ebbers bought another company, Grubman would put out a bullish research note, and the market capitalisation of WorldCom would increase by more than the price Ebbers was paying. Ergo, all WorldCom's acquisitions were successful. It was quite a nice little virtuous cycle while it lasted: super-charged revenue growth drove up the stock price and p/e ratio; a super-charged stock price gave Ebbers a highly valuable currency with which to make further acquisitions; and the fast pace of acquisitions drove the company's revenue growth.
All this was dependent on Grubman and his fellow telecoms analysts: it was they who never pointed out that WorldCom wasn't actually adding any value to the companies it bought, and that it made no sense to behave as though any company automatically doubled in value the minute it got bought out by the Ebbers machine. There was a good reason for them not pointing this out: while neither of the companies concerned got much in the way of bottom-line value out of the mergers, the banks who advised WorldCom on its acquisitions made hundreds of millions of dollars in M&A fees. (They got nearly as much out of underwriting WorldCom's ever-increasing debt issuance, as well.)
So there was never any doubt where Grubman's $20 million a year was coming from: it was trickle-down from the deals he was incessantly pushing, and the fees they generated for Salomon Smith Barney. Ebbers and Grubman needed each other, and both profited handsomely from the relationship. Investors signed on for the ride because while it was working it worked for them, too. If you followed Jack's picks, you'd make a lot of money. In the bubble years, stock prices often rose simply because they were rising, and not for any fundamental reason; the Grubmans of this world were there to reverse-engineer some sort of vaguely plausible rationale for the behaviour of an irrationally exuberant market.
So now we see why Michael Armstrong, the CEO of AT&T, would put pressure onto his friend Sandy Weill, the CEO of Citigroup, to get Grubman to upgrade his company. If Jack Grubman - long a thorn in AT&T's side - were suddenly to upgrade Ma Bell, all three of them could jump onto the ensuing bandwagon and make a lot of money. So Grubman takes another look at the company, suddenly decides he likes what he sees, upgrades it, and - presto! - Citigroup/Salomon Smith Barney gets a $45 million gig underwriting an AT&T spin-off. (A few months later, Grubman changes his mind and downgrades AT&T again, but by that point the money is in Sandy Weill's bank.)
Where Grubman and Blodget tripped up was in believing their own hype too much. For years, there had been a direct correlation between the degree of their own optimism and the degree of their own success. (Blodget made his name by putting a $400 price target on Amazon.com when it was trading at $243; Amazon hit the target less than three weeks later.) So when the market turned sour, they stayed on the bullish side, touting the same old stocks for the same old reasons, but not getting the same old response any more. They should have realised that the party was over, and that every virtuous cycle can turn into a vicious cycle. They should have realised that falling stock prices can be just as self-fulfilling as rising ones, and jumped onto the bearish side of the market. But they had been too optimistic for too long. (It was the fact that they were among the truest of True Believers in the first place that had led to their success.) So individual investors, along with Spitzer and Morgenson, can now blame them for maintaining "buy" ratings all the way down on stocks which lost 99% of their value.
We're now left picking up the pieces, in a world where investment banks are bending over backwards trying to disclose their inevitable conflicts of interest. But no amount of Chinese walls and disclosure statements will ever make analysts trustworthy again. Only when analysts' salaries drop back down to pre-bubble levels, and they're no longer superstars in their own right, will it make any sense to take investment advice from a sell-side institution.
Posted by Felix at 11:32 EST | Comments (0)
Smoking in Manhattan
Stefan Geens, a man whose description of his blog on nycbloggers.com starts with the sentence "free trade is good," has gone decidedly off Milton Friedman's deep end in his latest post. It reminds me of the old joke: How many Chicago School economists does it take to change a lightbulb? None: if the bulb needed changing, the market would have done it already.
In Stefan's conception, if 75% of New Yorkers wanted non-smoking bars and restaurants, then 75% of bars and restaurants would be non-smoking. Since this hasn't happened, then any poll which shows such a thing must be "a load of bollocks".
Before getting on to the reasons why Stefan's argument is a load of bollocks, let me point him first of all to the relevant press release from the New York City Coalition for a Smoke Free City, and also to the New York Times article (linked to from stefangeens.com, of all places) which says that 76% of the 7,000 members of the New York Restaurant Association favour a law banning smoking in bars and restaurants. Stefan wonders whether the former poll might not have made it clear that restaurants and bars were considered workplaces; the press release clearly states that the question was about "all workplaces, including offices, restaurants and bars".
In fact, it's easy to see both how 75% of New Yorkers might want smoke-free bars, Stefan's own personal opinion to the contrary notwithstanding, and also to see how the mere desire for such things might not be sufficient to bring them into existence.
First things first: Non-smokers like smoke-free bars for reasons which everybody knows. But if you go to LA, where smoke-free bars are the law, you'll even find a lot of smokers lauding them. Were they common before the law was passed? No. Here are some reasons why not:
- The relative force of the prohibition. If smokers aren't allowed to smoke by law, they won't. But if an individual bar decides to go unilaterally non-smoking, it's likely to be on the receiving end of a lot of smokers' complaints.
- The fact that most groups of friends include smokers, who are hooked to a highly addictive substance; furthermore, their addiction seems only stronger when it's combined with drinking. In short, smokers' desire to smoke in bars is a lot stronger than non-smokers' desire to drink in non-smoking bars.So a group of smokers and non-smokers will inevitably go to a smoking bar if given the choice. Only if the smokers have to step outside to light up, will they.
- The fact that Beautiful People, the trend-setters with the white American Express cards who most bars want to attract, tend to be much more likely to smoke. And non-smokers, who have lived all their lives in a city with smoking bars, have shown themselves more than willing to go to such places. So switching from smoking to non-smoking status carries quite a large downside (the loss of the smokers and their friends) with very little upside (the extra patronage of the boring types who seek out non-smoking bars, should such people actually exist).
Basically, it seems as though Stefan has confused the desire for New York to ban smoking in bars with a desire to go to non-smoking bars, given the choice. It's a subtle difference, but an important one: I'm sure that most of the 75% would rather go to a cool bar with smoking than an uncool bar without it. But they'd prefer even more to go to a cool bar without smoking, which is only going to happen if a law is passed. That's why they want the law!
Personally, I'm agnostic on the issue of whether I prefer smoking or non-smoking bars. I've been to good exemplars of both. But the way Mike Bloomberg is presenting this, my own preference is not really the point. If I don't want smelly clothes and red eyes, I don't need to go to bars in the first place. Bloomberg's point, which I think is unarguable, is that bar proprietors and staff are seriously harmed by second-hand smoke: his proposed law would exist primarily to protect them.
All manner of workplace-safety laws are on the books already, for very good reason. The way I see it, this is another law of that ilk. Bartenders have to work in the smokiest sections of the smokiest bars for eight hours a night, up to six nights a week. That's a horrible workplace condition, even if the bartender is a smoker. I predict that if this law goes into force, there will be a lot of grumbling for a few months, and then people will realise that they actually prefer it the new way, while bars will realise that they've lost no custom. The population will be healthier, bartenders will be happier, and everyone will win. Especially the bar owners with gardens.
Posted by Felix at 11:38 EST | Comments (5)
The World's Greatest Atlas
More traveller’s thoughts: this time one of those ideas you have when you wake up in a strange hotel room at some time of day when you’re really meant to be going to sleep, not waking up, and you’ve just had the weirdest dream about old Ordnance Survey maps, not of the UK but of the world, and they fit together onto a big table, six of them, side by side, and they’re relief, with the Amazon rainforests burning real flames and the Gulf Stream shown in the Atlantic in real blue water, and it’s all dated circa 1955 or thereabouts, and you’re trying to get people, including your ex-boss, to see this amazing old thing which you’ve discovered on the shelves of a second-hand bookshop which is packing up to move elsewhere, but it’s down the end of a dark alleyway, and it’s really hard to make them interested – you don’t know what I’m talking about? Never mind, this is great idea anyway. It’s not original, I’m sure, but I’m loving thinking about it, and because this is my website, I can put up on it whatever I want.
So… The World’s Greatest Atlas. As we all know, maps are all computerised now: it was a lot of work to take all of the information which had painstakingly been immortalised in print and transfer it into a huge relational database, but now it’s been done, and both Bertelsmann and the OUP now have incredibly detailed information about pretty much all of the surface of the earth, which they can print out in various different forms for different atlases aimed at different markets all over the world.
So: put it all on a DVD-ROM, or something like that: I have no idea what the ratio is of the sort of information I’m talking about here to the storage capacity of a multimedia disk, but if it can’t be done, then just put it all on a big fast web server (actually, come to think, that’s even better, because then all the information is kept up-to-date in real time) and sell the software which is required to read it over the high-bandwidth pipes we’ll all have in next to no time. I’m not joking here: this could be the high-bandwidth killer app which finally gets people to upgrade from dialup en masse, and which also manages to be the first website outside the business information and pornography industries which millions of people are actually prepared to spend money for. (The economics are great: you can have annual subscriptions, or just pay on a per-visit basis; the kind of things you can find on the internet anyway can be free, but higher levels of granularity can cost more, that sort of thing. Hell, if it gets big and fabulous enough, you might even need some sort of CIA clearance to get to the really detailed stuff!)
Now the software’s the real beaut. Pick up any world atlas in your local bookshop, and you’ll see that the maps in most of them are hideously garish. Maps used to be things of real beauty, but now most publishers don’t have the resources to make beautiful maps any more. They just hit a button on their relational database, pick a few colours, and let it fly; some peon in the graphic design department then spends maybe a couple of hours on each one making sure that the place names don’t overlap too badly, and it’s off to the printers. The problem is, the maps have to show far too much information in far too little space: they have to be all things to everybody. Someone looking up a small town in south-western Germany has to use a map showing the topography of north-eastern France. With a computer generating maps on demand, however, all of that is a thing of the past. If all you want is the cities of south-western Germany, that’s all you’ll get. If you want a general topography of western Europe without bothering with lots of useless place names, you’ll get that as well. Everything can be done to the scale you want, with only the information you want. The mapmaker’s art is that of fitting lots of information into an enclosed space: this software will do away with the mapmaker’s art (to be honest, it’s pretty much dead already anyway) by having much less information in an essentially unlimited space.
The latest edition of the Times Atlas of the World did away with the city maps. On the one hand, you can see why: they’re certainly of no use compared to the sort of city maps you can find for free on the internet. But at the same time, it’s a real loss to the atlas. Our new software can drill down from the world to the Americas to North America to the United States to New York State to New York City to Manhattan to – and this is where New York City’s own map gets integrated into the atlas – the very block you live on, with its water supplies, its buildings, street numbers and everything. You can even see it in photographic form if you want.
This atlas will do things no one has ever been able to do before: pull up a topographic map of the world, say, and then at the touch of a button evaporate all the water: see the surface of the earth without the arbitrary cut-off at sea level which most maps make. With the three-dimensional data available, you won’t even need to confine yourself to the standard bird’s-eye view: you can move around the canyons of the earth and sea just like you can move around an unbuilt house using a CAD program.
Political boundaries will be constantly updated, of course, but the old ones won’t be erased: type in a year, and the political map of the world for that year will immediately appear. Press a button, and you can fast-forward through wars and treaties and see Europe’s states alter over time, watch the African independence movements slowly appear.
You want a road atlas to get you from Peoria, Illinois to Nashville, Tennessee? You’ve got it. You want to see where the world’s known oil reserves are? Here you are. You want a map of caribou breeding grounds in the Arctic National Wildlife Reserve? Pronto. Add in data from space satellites, and you can see a pictorial representation of just about anywhere on the earth’s surface, as it looks right now.
We’re not there quite yet, of course, although we’re not that far away. The great thing about this atlas is that it doesn’t need to be up to full strength immediately: just the ability to manipulate the information available right now would be something incredible. There’s nothing the software needs to do which hasn’t already been written in some form. Getting the information together in one place and in one format is the tough bit, sorting out copyright issues, that sort of thing. But this isn’t just a cartographic version of the Humane Genome Project I’m talking about here: its usefulness far outstrips the relatively small world of maps and mapmakers. There’s something in it, literally, for everyone. Please let me live to see it!
Posted by Felix at 1:35 EST | Comments (0)
RSS - All blog entriesFelix Salmon: Recent posts
Felix's del.icio.us links
Archives
This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 United States License