A few posts ago I wrote a bit about the long term, particular in regards to it being little more than a series of short runs and this generally being the best way to look at it.
This view did of course neglect at least one important point, the need for some long term investments. Obviously, things like infrastructure and education have very high initial costs and fairly low year on year returns. They are distinguished by having very long periods over which a return is generated.
My views on the long term being unpredictable aren't meant to argue against this kind of investment. Though I do tend to view them a bit differently from a standard cost benefit analysis would. I see these investment's purpose primarily as providing long term flexibility in a changing world. We can't predict in the long term what will be the great growth areas of the future, though we can predict the stable growth areas, it's quite predictable that decent returns will be generated from improving infrastructure in NYC for instance, however, since NYC is already quite wealthy it is impossible to generate the kind of massive returns that were generated in the past in Silicon Valley, Las Vegas, or Phoenix, the highest rates of return are always in areas that are relatively underdeveloped to begin with and creating the conditions for this kind of serendipitous rise is what I see as the primary reason for long term investments. Private, medium term investments will suffice to develop long run growth in a place like NYC or to educate the upper management of the future, what it won't do is provide us with the unpredictable rise of formerly remote regions or the mass of nameless but proficient middle managers, technicians, and scientists necessary to make things work in a modern society.
Of course, this means a great deal of waste. But so does private investment, creative destruction is a necessary component of private investment because individuals often make mistakes with their allocations of resources. Wasted long term government expenditure is little different. For every success, there will be numerous boondoggles, that is the way of any enterprise in this world. With government, these projects are always readily visible, in the private sector, we often are blinded by survivors bias, we often forget about how much failed investment occurred alongside every Amazon. It is unreasonable to expect the government to do better.
This also of course doesn't fix the problem of unpredictability, think of canal building for a huge amount of wasted long term investment. The essential quality of this investment isn't that it will accord well with cost-benefit analysis however, it's that when it pays off it raises the ceiling on total available wealth and increases access to productive activities, having unpredictable effects. Just because we have no real idea over whether or not it will be worth doing doesn't mean we shouldn't do it, it just means we have to admit we're still basically shooting in the dark with these projects. I consider madness to look at uncertainty as a reason to simply fall into senescence, no matter how emotionally compelling sloganeering about uncertainty is.