I started dealing with the rôle of construction in economic development in the late Eighties. By the early Nineties I spoke at several conferences and published a paper on this topic. Considering that industrialization and urbanization are closely bound, as well as that manufacturing and construction are close partners in development, I hypothesized that the share of construction in total output was low at early stages of economic development as measured by output per inhabitant of a country, that it rose with growth, and that it ultimately declined at late stages of economic development. The time horizon I had in mind was about a generation, or twenty-five years. The diagram I drew to depict this relationship was deliberately of the simplest kind so as not to imply any specific mathematical relationship. To this end, I used a segment of a circle. For simplicity, I referred to it as an inverted U-shaped relationship.

At the time, my hypothesis was rather bold. The standard view even in the Nineties was that the share of construction in total output was constant throughout development. Given the statistics from the Sixties and Seventies, this made some sense. As an economy grows, all sectors grow. And so does construction. This was expected to hold for ever and ever. But my hypothesis was based on the first signs of change in the Eighties, and especially in highly developed economies. Once the construction activity started sagging, this needed an explanation. And I provided one that made most sense in view of what was known about economic development in general.

Over time, my diagram turned into “the Bon curve” in the literature. This is how it is referred to in all sorts of papers and books. I remember seeing one of them that claimed that the relationship between the share of construction in output and economic development was inverted V-shaped for a particular country rather than inverted U-shaped. Fair enough, assuming the country is small enough. But I have just received another paper on the topic that is being considered for publication by a respectable journal. It purports to show that even my inverted U-shaped relationship is wrong. There is a great deal of statistics to show this to be the case. Of course, the last date for which relevant statistics are available is just before the onset of the greatest economic crash since the Great Depression. And the boom preceding it was caused by irrationally-exuberant finance, real estate, and construction, of all things.

Now I am wondering what to do about the threatened Bon curve, though. After all, it caries my name. But I am not interested in joining the fray at this stage of my life. I do not even have a single paper or book I have written about this topic. Besides, I have better things to do. On the other hand, I cannot trust my former colleagues to defend me, either. Many of them are retired or close to retirement, too. Perhaps the best way forward is to let things be and trust that the new generations of researchers will understand my hypothesis as it was intended—as a general proposition about the rôle of construction in economic development. The share of construction in total output cannot but drop at late stages of economic development, when new construction is gradually replaced by maintenance and repair. And the time horizon can be extended at this point: it cannot but be as long as this civilization lasts. Long live the Bon curve!