The trouble with market reporting

Barry Ritholtz hits upon a bugbear of mine - the atrocious reporting of daily swings in stock markets. He says: "Pat explanations for the markets’ daily gyrations rarely ring true."
He's right of course. But I'd go further. They are often nonsense - merely a search for a pattern in what is really random data.
For example, today the FT tells us that "strong gains on Wall Street overnight were fuelled by easing oil prices." The Times says" "Wall Street was boosted overnight after the oil price eased and after an upbeat report on consumer confidence."
It sounds plausible. But it runs into a problem. So far this year, the correlation between daily moves in the S&P and oil prices, though negative, has been insignificantly different from zero. The correlation between weekly moves has also been statistically insignificant, but positive. That's what you'd expect in noisy data.
Explaining market moves are just easy after-the-fact rationalizations. In weak macroeconomic  conditions, a market fall will trigger the line: "market falls on economic fears."  And a  rise will invite: "market rises on interest rate hopes." It's the same macroeconomic story.
There are several clues to this pin-the-tail-on-donkey approach. One of the best is "after". It hints at causality without stating it. Why not say: "market rose today after I missed my bus"?
Paradoxically, of all the explanations the dead trees give for market moves, they never offer two of the most likely ones. You never see the stories: "market rises but it's just random noise." Or "market rises because risk aversion declines and, hey, de gustibus non est disputandum."
Market reporting isn't just stupid, though. It is potentially downright pernicious. In offering a spurious reason for every blip in the index, such reports invite us to believe there's a mechanical link between economic conditions and share prices. They therefore encourage naive investors to believe the market is safer than it really is. Worse still, they encourage them to trust the "experts" who seem to understand such links.
The fact is, though, if you want to understand financial markets, you must ignore (almost) everything you read in the dead trees.

Comments

[dearieme]

Oh, who'd believe Pat anyway?

[John East]

Can I also comment on the often extreme positive spin perpetrated by the financial media. I don't read the financial press very often, so my observations may, or may not apply to them. I am referring mainly to the commercial broadcast media such as Bloomberg and CNBC. For example, falling stock prices are frequently referred to as "volatility". In fact the abuse of this word has got so bad that it may well end up defined as "a fall" in future dictionaries. The spin continues on days when shares rise. Commentators and interviewees are wheeled in front of the camera to proclaim the start of a new multi-year bull market etc. etc., and we are all told to rush out and buy value stocks such as Google.
I suppose a lot of journalist's jobs depend on everybody remaining in euphoric spend mode, but when a turndown comes, which must happen sooner or later, the size of the financial collapse will be much bigger than it needed to be thanks to these journalists.

[Toni]

Having written a gazillion of market reports for a newswire let me add my 2 cents.
The appreciation of after-the-fact reports rises equally with the distance of its readers whereas the local sources for the very same reports want to read forward looking statements, i.e. what will move it and in what direction.
Silly as they sound very often they are still the most read reports, access statistics have shown. Sometimes they help the price spiral. It happens very often that the same sources that were phoned up for the reports use them in their in-house and customer fact-sheets, saying, "see, this wire says also the market will move that way".
The reasons given for day-to-day moves are sometimes shallow on the surface, but who would read a report that states day after day, "the fundamental outlook has changed marginally, but stocks still moved." No filing editor would accept such an un-sexy report.
Trader's comments are to be taken with caution anyway as they see a call for a market comment as an opportunity to talk their book.

[Angry Economist]

Forecasting or even something more basic, trend analysis and projection, can't be accurate enough to predict and explain the minutia of events or decision making. The best it can do is tell you what is likely to happen, maybe, perhaps, given quite a few caveats. And its poor at explaining post- or ex- ante, why things occured or will occur. People tend to rationalise and put their own explanations around trends and numbers to justify things. To properly explain why things happened involves a lot of research and analysis. Most stock mkt journalists don't do this. Most good financial analysts don't publish this - its their tacit skill that they get paid big bucks for.

Governments do exactly the same thing as stock market journalists. In the government's case its peddling the mantra of skills and the 'knowledge age' etc etc.

My experience of forecasting and research - people want the be told what to do and are essentially lazy and too trusting. They definitely don't want to analyse the situation for themselves.

The worst case is when their own 'mental model' of the market becomes too entrenched and they are unwilling/incapable to change their market position, and its faultiness is cruelly exposed by market events.

Mind you the financial services industy is quite adept and manipulating human psychology in this way, just like politicians are!

I am reminded of this! MONORAIL, anyone!?
http://origin.thesimpsons.com/episode_guide/0412.htm