Why GM can only get hurt keeping data secret

General Motors won’t release production data?

Red alert! Like when the “Check engine” light pops on, the immediate question is “How bad is it?” You hope it’s a $20 fan belt but fear a cracked engine block.

As Automotive News reports today, GM has decided to stop publicly disclosing monthly production numbers. That’s instantly alarming to economists, analysts and GM’s suppliers. Number crunchers rely on the GM data for their own forecasts, and suppliers for their business and production plans.

Let’s get some perspective. GM isn’t legally required to report the data. But it should because keeping the data secret works against GM.

First, it violates public trust. Almost all automakers report their output. The industry, investors and the public are used to it and expect it to continue. Now GM expects to see competitors’ numbers but keep its data secret.

Second, it reverses decades of increasing openness. When I worked for Ward’s Automotive Reports in 1972, Editor Harry Stark told me about the bad old days of 1930s data collection. Every Friday, Harry and his boss drove around metro Detroit handing $50 bills to assembly line foremen for that week’s count.

“The automakers howled we were selling trade secrets to their rivals, but somebody finally asked who was subscribing. We said, ‘Your suppliers, who want to do a better job.’ ‘Oh.’” Within months, automakers provided actual production numbers.

Third, withholding information sets a bad precedent and raises doubts.

That uncertainty will directly increase GM’s operating costs. The more uncertain a supplier is, the more margin it builds into bids. The more uncertain a dealer is, the fewer units he wants to commit to and the less likely he is to upgrade a GM store. Uncertain bankers could charge higher interest rates. Uncertain consumers are less likely to buy GM vehicles and less likely to pay as much if they do buy. And uncertainty alarms investors, driving down GM stock prices.

Finally, it won’t work anyhow. Production information is available piecemeal. Suppliers count the parts they make. They talk to data providers. But data collection will cost more for fuzzier numbers. If you force industry folks to estimate, they forecast low.

This isn’t the 1940s, when Ford Motor Co. used to estimate next month’s cash needs by bundling its invoices into a pile it physically weighed on a truck scale. Today’s managers base their decisions on quantifiable data -- some even call themselves quants -- and will pay extra to fill data holes.

Because bad data can be costly. In Europe last decade, automakers had started reporting sales and production through their trade association, ACEA. But the data didn’t smell right, especially the notion that European dealers self-registering thousands of new cars monthly actually sold them to individuals as “zero-kilometer” used vehicles. The dirty secret: Dealers did sell those self-registered cars, but usually to dealers in other countries, who then resold them as new.

If a car is shipped from Italy to Austria to Slovakia to the Czech Republic, how many times is it reported as a new-car sale? Suspect data were more a symptom than the cause of the ensuing crisis but certainly didn’t help decision making.

In my experience, if a corporation starts concealing information, it’s either lying or preparing to do so. Wall Street makes the same assumption: Any withheld info is bad news.

And here’s the worst part. GM knows every bit of this -- and chose this path anyhow. Good golly, just how bad is the situation?

GM just shot itself in the foot. Before it further damages its suppliers, dealers, employees and investors, it should resume reporting monthly production.

You can reach Jesse Snyder at jsnyder@crain.com -- Follow Jesse on Twitter: https://twitter.com/spartyjesse