This morning, while riding the express train to work, I stood facing one of those ubiquitous census ads and, for the first time, began considering its wording in earnest. I am sure you’ve seen it too: “If we don’t know how many schoolchildren we have, how can we know how many schools to build? … If we don’t know how many people we have, how can we know how many hospitals to build?” And so on and so forth.

That the government should still pose such questions — innocent as they are — suggests that the so-called problem of economic calculation afflicts the endeavors of central planners today no less than it did in the 1920s, when Ludwig Von Mises first set it forth. Not only that, but the government has also failed to find tools more efficacious in tackling this problem than the nationwide survey — that is, the census. And what a crude device that is!

For one thing, any information collected through it soon becomes outdated, since the census is taken at intervals of no less than 10 years, during which time a lot can happen in terms of economic development and population shifts. For another, delivering the surveys to every doorstep in the country, entreating the citizens to fill them out, and ensuring that a tolerable number of them actually do so amounts to an onerous affair not cheap to orchestrate — as is plainly evinced by the budget of $11.3 billion set aside for the accomplishment thereof. And for all the pains that go into collecting it, this information winds up reaching the government incomplete and only approximately accurate — the proportion of falsified surveys that alloy the census results being a matter of contentious and largely partisan debate.

Numerous and intractable as these difficulties are, they don’t even touch on the core inadequacy of the census at informing central planning — which is that no questionnaire could possibly gauge supply and demand. Only the market can take their measure, through prices and, to some extent, interest rates — its two chief signalers. It is by dollars that people vote for what they want and how much of it they want and how badly. The results of such “elections,” the likes of which are held daily in the marketplace, might differ widely from anything obtainable by aggregating the individual wish lists of the “voters.” And because, on the whole, they reflect practical reality as opposed to vague or impossible fancies, the former results are always preferable to the latter and form the only sound basis for business planning. This is the crux of the problem of economic calculation: the reason why in every kind of regime past or present, even the most well meaning of central planners have found it impossible to allocate scarce resources efficiently throughout the economy.

To the not-so-rhetorical questions of the census ads, such as, “If we don’t know how many people we have, how do we know how many … (fill in the blank) buses we need?” one all-purpose answer comes to mind: you, the public sector, cannot know how many buses a line needs, or how many schools and hospitals should be built in an area, and knowing how many people live nearby might not even help you to find out. But the private sector eventually gets those answers right. If there are too many buses in a line, the private transportation company incurs a loss and cuts some of them. If there are not enough, someone will endeavor to introduce one more and thus realize a profit. The same argument applies to hospitals, clinics, and schools. Leave the matter to the businessmen and entrepreneurs, whose livelihoods depend on it. And don’t employ the census toward any end except the one dictated by the Constitution — i.e., to apportion the number of members of the United States House of Representatives among the several states.