Musing about the Future

Musing about the Future
MiBiz • May 31, 2004
by Jim Gillette

“Plans are nothing; planning is everything.”

– Dwight David Eisenhower

I was recently asked by a client to provide a detailed global annual vehicle production forecast out to 2015. We normally provide forecasts only five years into the future; eleven years, after all, is a very long time. I nevertheless look upon such an infrequent assignment as an exciting challenge well worth the effort.

As I see it, there are two good reasons for thinking about the future. First, you don’t want to bet your company on assumptions that later turn out to be boneheaded. Second, innovation is born of thinking of different ways of solving problems under alternative scenarios.

Five years out is about as far as you can go if you expect a relatively high level of accuracy at the manufacturer’s platform or program level. Auto industry forecasters looking at mature markets like the United States are helped by the fact that nearly ninety percent of vehicle purchases are repeat buys. There is enormous data available to track consumer trends and preferences that to a large extent carry forward from purchase to purchase. Trends are relatively easy to discern.

The problems start when some event knocks the market off the trend. With more than 57 million light vehicles assembled around the world every year in over a thousand plants supported by tens of thousands of parts plants, the global auto industry is a complex system. Complexity theory (or popularly known as Chaos Theory) postulates a small change in the initial conditions of a system can result in unpredictable outcomes over time. The statement that a butterfly flapping its wings in Beijing results in a thunderstorm in New York two weeks later is often used to illustrate the point.

A forecasting model I built in 1990 contained 83 variables that affected the outcome of a given auto company’s build rate in any given calendar quarter. A significant change in any one of the variables could affect demand over a period of months and years, yet the actual results cannot be forecasted with precision. Interaction between the variables and what is know as two-way causation makes such large scale multivariate models undependable as forecast tools for periods beyond a few months. (Large-scale econometric models built to forecast the U.S. economy have been out of vogue for years because of their failure to accurately predict turning points in the economy.)

The main benefit to thinking about the future, however, is not that you will be able to predict market demand, consumer preferences, or commodity prices with any degree of accuracy. The payoff comes in identifying and testing the assumptions you currently hold about your business and its relationship to the markets you serve.

Futurist Joseph F. Coates points out, “The central feature of all organizational failure is that an individual or a few people at the top had assumptions about the future that were unsound.”[1] The process of studying the future forces us to carefully examine the assumptions we hold about the future and to test their rationality and likelihood of being correct.

The global auto industry is littered with examples failures resulting from faulty assumptions. Given the recent rise in oil and gasoline prices, it might be instructive to look at one case that occurred during the early 1970s when energy prices first became a problem for the auto industry.

The then Hydramatic division of General Motors had been working on a front-wheel-drive transaxle for use in a future small, fuel-efficient compact car. Assuming that $2 per barrel oil would be readily available for the long term, GM decision makers saw no need to develop and market a fuel-efficient compact in the US market. The transmission was cancelled well before the oil embargo of 1973. When the oil crunch came, GM lacked a product for the time. The only available option was to gear up production of a Brazilian made rear-wheel-drive subcompact, the Chevette. To put it kindly, the result was less than optimal and the Japanese competitors had their entrée.

Not quite two decades later, very few automakers had positioned themselves to take advantage of the surge in SUV sales in the US. In fact, I can find no record of any industry pundit foreseeing the trend as of 1989. In a somewhat rare occurrence, Ford outsold its original sales forecast of the new Explorer by about 40 percent the first year, and this was during the recession surrounding the first Iraq war when gas prices spiked upward.

In hindsight, it is evident now that both oil and gas prices had been running well below their historical trend since the temporary collapse of OPEC’s ability to manage the market in 1985. The decline of the relative fuel prices proved to be a powerful factor enabling SUV and other truck products to gain a foothold, most recently comprising more than 50 percent of US vehicle sales. Coincidently, Americans demanded more powerful engines, driving average fuel economy from its high in the 1987 model year of 22.1 mpg to 20.8 mpg in 2004.

Is the current trend in fuel prices a prologue to the future? Other than to state that oil and gas prices are likely to be higher in 2015 than they are today, I won’t venture a guess. I knew a lot of Texas business owners who thought high oil prices were here to stay just before they dropped from $30 per barrel in 1985 to just over $12 the next year. The most serious mistake non-economists make when attempting to predict long-term trends is to ignore the self-correcting mechanisms of market economies.

So why is it so important to develop a forecast of the global auto industry out to 2015? We don’t want to miss the obvious. Business investments are made based upon assumptions that too often get baked in. Both physical assets and mindsets become dedicated and fixed rather than adaptive and flexible.

Clayton Christensen at Harvard and other researchers studying innovation have demonstrated time and again that established leaders in an industry rarely produce what Christensen calls destructive innovation; that is, changes that are so fundamental that old business models are completely replaced. Newcomers, unconstrained by sunk costs and statements like, “We’ve always done it this way,” create market-capturing breakthroughs. A few minutes reflection by anyone familiar with the auto industry should bring numerous other examples to mind.

Projecting ahead to 2015, if your company has achieved or maintained market leadership, you might ask yourself, “Was it luck or is it preparation?” It probably won’t matter to you then, but I wouldn’t count on luck.



[1] Coats, John C., “Why study the future?”, Research Technology Management, May/June 2003