Forecasting

We are constantly presented with forecasts, be they economic or otherwise. While it is pleasant to speculate on the future, we often overestimate the usefulness of forecasts which are mostly stabs in the dark even if cloaked in the appearance of science.
Recently a friend showed me an interesting document that I think shows how far off a forecast can be. It is titled The Soviet Space Challenge and was published by the US Department of Defense. It includes a preface by then Secretary of Defense, Caspar Weinberger dated November 1987.
Students of history will recall that the Berlin Wall was pulled down in 1989, and the Soviet Union formerly ceased to exist in December 1991. I have copied a chart from the publication titled “Soviet vs. US Weight to Orbit” which shows an explosion (red line) in Soviet space capability from 1985 to 2005. The text states (p.20):

Clearly, the Soviets understand the long-term strategic importance of space. Their current and planned space programs and their developing capabilities are structured for maximum future exploitation of space operations. The chart on page 19 [below] clearly illustrates that between now and the year 2005 the Soviets will deploy a tremendous weight-to-orbit capability. Their identifiable launch requirements, we estimate, will be two to three times our own, while their projected launch capacity between 1990 and 2005 is nearly double any requirement that we can presently identify. This raises the possibility that the Soviets are already considering the lift requirements necessary to expand rapidly their large-scale military presence in space.

The lesson? Take forecasts and projections with a grain of salt. Be flexible so as to adapt to changing circumstances. Be prepared for events to unfold in ways not expected, anticipated or even considered improbable.

This entry was posted in Uncategorized. Bookmark the permalink.

18 Responses to Forecasting

  1. Biota

    Pascal said in Pensées,if I recall correctly

    the present is their means, the future is their end, therefore they do not live.

  2. Jim Rose

    Macroeconomic forecasting has had a turbulent history.
    Most early discussions argued against econometric forecasting in principle.
    • Forecasting was not properly grounded in statistical theory,
    • It presupposed that causation implies predictability and
    • The forecasts themselves were invalidated by the reactions of economic agents to them.
    A long tradition argued that social relationships were too complex, too multifarious and too infected with capricious human choice to generate enduring, stable relationships that could be estimated.
    These objections came before Hayek’s point that much of all social knowledge is not capable of summation in statistics or even language.
    In the 1980s, data mining and publications bias were so strong and statistical inferences were so fragile that that Ed Leamer’s 1983 Let’s take the con out of econometrics paper made up and coming applied economists despair for their professional field and for their careers.
    Leamer still doubts the progress towards techniques that separate sturdy from fragile inferences. Economists by and large don’t want to hear that they cannot make major conclusions from the data sets.
    Before the great moderation spread wide, Brunner and Meltzer found that the 95% confidence intervals on next year’s forecast for Gross Domestic Product and the Consumer Price Index are such that government and private forecasters in the USA and Europe cannot distinguish between a recession and a boom, nor say whether inflation will be zero or ten per cent.

  3. Rafe Champion

    For many years Paul Samuelson wrote in his best-selling textbook that the Soviet economy was growing strongly and rapidly overtaking the US. Those were the days!

  4. Jim Rose

    Palin is polarising with little appeal to the independents that decide elections but so was Reagan, but he was older, wiser and had five successful careers behind him: radio sports-caster, movie actor, trade union president, corporate spokesman, and two-term governor

  5. Tim Quilty

    Just as well the Soviet Union fell over before they achieved that dominance in space or we might all be stuck with communism now instead of our slowly creeping nanny state…

  6. Don’t trust any forecast which doesn’t come with a confidence interval. Without it, the forecast is not worth the paper it’s written on.
    This goes ‘a fortiori’ for Treasury and Bank forecasts.

  7. m0nty

    Now that’s a hockey stick!

  8. ben

    this is clearly someone talking their own book. we need more cash because look at what the Soviets are about to do!

  9. .

    Zandi uses a 900% forecast error.

  10. Wouldn’t it be more reasonable to expect that the author (US dept of defence) wanted the reader (politicians) to fear the expansion of Soviet space capability to increase US funding to space exploration, and indeed themselves (defence)?

  11. CraigS

    Kind of like Treasury forcasting that 730,000 people would open first home buyer accounts, when blind freddy could see that was a joke.
    Actual number to date 23,000, just a touch off.
    The real scary thing is that governments make decisions on this sort of stuff!

  12. FDB

    No Cameron, it’s cos models is always rong an stuff.

  13. .

    Treasury had a 2600% forecast error? Hilarious.

  14. Rafe Champion

    Re Cameron Murray comment, the CIA also reported that the Soviet economy was going like a rocket, no doubt for the same reason.

  15. Rafe Champion

    As Paul Keating used to say, in the race of life, put your money on self interest, at least you know its trying!

  16. Jacques Chester

    One thing I’d like to see spread out of software estimation is the “cone of uncertainty“.

  17. TBH

    I’ve come to the view that much macro punditry and forecasting out there, particularly from high profile individuals, is nothing but talking one’s own book. Guys like Jim Rogers, Bill Gross, George Soros and Jeremy Grantham are among the worst offenders.
    I’ve stopped listening to pretty much every one of the above and plenty of others. None of them have a particularly clear crystal ball.
    About the only guy I read with regularity these days is Don Stammer, as his commentary is usually pretty measured and relatively free of hyperbole.

  18. JC

    Guys like Jim Rogers, Bill Gross, George Soros and Jeremy Grantham are among the worst offenders.

    Mr. Bow Tie is the worse… Ever seem him (Jim Rogers) interviewed on his treadmill?

Comments are closed.