Let’s put forecasting in its place

Make text smaller Make text larger

  • Beware fortune tellers: the crystal ball is often defective

    Beware fortune tellers: the crystal ball is often defective


One of the things I failed to learn after writing a book was the simple fact that methodically demonstrating a thing with data and evidence doesn’t resolve that issue. I shouldn’t have been so naive, given that some people still are flat-earthers, or are anti-vaxxers, or Holocaust deniers, or claim that global warming is a Chinese hoax.

Which is why, despite all of the earlier discussions about the folly of forecasts, I find myself once again compelled to bring up this subject. Blame it on the time of year, when all of the forecasts for 2017 are being rolled out, while the old ones that were so-often wrong are forgotten instead of being reviewed.

So please consider this column a public service. Here is a round-up of some of 2016’s forecasts, and a look at why they didn’t quite work out as expected.

Get Ready for $80 Oil: “When oil drilling activity collapses, oil supply goes down too.” That may or may not be true, but one thing we know is that oil never got anywhere near $80 this year, peaking at about $53 earlier this week. The error here is assuming you can accurately measure a drilling activity slowdown; figure out what that impact will be on supply; and determine how that will affect prices. This forecast, made by Raymond James, tried to thread that needle. It didn’t happen.

The recession of 2016: “Central bank bungles, oil price fluctuations and overregulation indicate contraction.” This forecast, published in the reliably ideological Washington Times, commits the classic analytical error of infusing emotional politics into forecasts. The author opposed the Federal Reserve’s programme of quantitative easing as well as government regulations, so of course these things will cause a recession! Only, they didn’t. Economic analysis coloured by political bias equals awful investing advice.

Recession sign is in play and has 81 per cent accuracy: recessions have followed consecutive quarters of corporate earnings declines (whenever that happens) four out of five times, according to JPMorgan Chase & Co strategists. There are at least two reasons for this: first, as we have discussed before, relying on a single variable to explain a complex system such as the economy is a poor approach to analysis. The world is just not black and white. Second, it confuses correlation with causation.

Billionaire Sam Zell says recession likely in next 12 months: this one blames the Fed and the strong dollar, which fell in the six months after Zell’s prediction. We have discussed Zell’s call before, but a few words about the halo effect, or the tendency to give people successful in one field more credibility than is warranted in other areas. Although we have to give credit to a billionaires’ ability to make money, they probably don’t have a comparable ability to foretell recessions.

Why gold will see $2,000: gold bugs are too easy to debunk. But as a reminder, the market has managed to figure out that China and India are net buyers of gold, and have been for thousands of years. That isn’t what is going to drive its price, which never reached more than about $1,375 this year.

“Wall Street fears a Trump presidency: stocks may lose 10 to 12 per cent of their value if he wins the November election, and there may be a broader economic downturn.” So wrote Justin Wolfers in the New York Times on September 30, based on a “close analysis of financial markets during Monday’s presidential debate.” This looks like another classic error of causation and correlation. This prediction, for all we know, may yet prove true. But we have to note that since the election, the Standard & Poor’s 500 Index is up 6 per cent.

This Dow rally will end March 23: this forecast was made on February 16, so at least it didn’t have much shelf life. Studies have shown that people prefer false precision to accurate ambiguity. The forecaster who tells you that the future is inherently unknowable, and all forecasts 12 months out are just guesses isn’t as well received as someone saying with certainty that the Dow Jones Industrial Average will top 21,000 on the day Donald Trump is inaugurated as president. Why don’t investors recognise this? The answer is quite simple: it is just human nature.

Those who make forecasts for a living, or are asked to do so by the news media, should use the opportunity when they make a prediction to point out the absurdity of the exercise. My own 2017 predictions attempt to do just that.

The bottom line remains: forecasts and predictions are exercises in marketing. Outrageous and wrong forecasts are typically forgotten, and when one randomly happens to come true, the guru is lauded as the next Nostradamus. It is an expensive and fatuous practice, and the finance industry should give it a permanent rest.

Barry Ritholtz is a Bloomberg View columnist. He founded Ritholtz Wealth Management and was chief executive and director of equity research at FusionIQ, a quantitative research firm. He blogs at the Big Picture and is the author of “Bailout Nation: How Greed and Easy Money Corrupted Wall Street and Shook the World Economy.”

You must be registered or signed-in to post comment or to vote.

Published Dec 17, 2016 at 8:00 am (Updated Dec 16, 2016 at 7:15 pm)

Let’s put forecasting in its place

What you
Need to
Know
1. For a smooth experience with our commenting system we recommend that you use Internet Explorer 10 or higher, Firefox or Chrome Browsers. Additionally please clear both your browser's cache and cookies - How do I clear my cache and cookies?
2. Please respect the use of this community forum and its users.
3. Any poster that insults, threatens or verbally abuses another member, uses defamatory language, or deliberately disrupts discussions will be banned.
4. Users who violate the Terms of Service or any commenting rules will be banned.
5. Please stay on topic. "Trolling" to incite emotional responses and disrupt conversations will be deleted.
6. To understand further what is and isn't allowed and the actions we may take, please read our Terms of Service
7. To report breaches of the Terms of Service use the flag icon

  • Take Our Poll

    • Has the Progressive Labour Party done enough to warrant a return to government?
    • Yes
    • 18%
    • No
    • 77%
    • Undecided
    • 5%
    • Total Votes: 5386
    • Poll Archive

    Today's Obituaries