After reading Nassim Nicholas Taleb’s book The Black Swan, I was much impressed, as has been nearly everyone who has read it. And I hastened to rethink my forecasting work in the light of the potential for the exceedingly rare, but vastly important events (“outliers” in more common parlance) that he named “black swans.” But in the end, I found little of relevance to the public policy arena. But let me explain. Among other duties, I forecast child care enrollments and costs for the State of Maryland’s State Department of Education, the lead agency for the Child Care Subsidy program locally. This program provides subsidies to low income families to help them afford safe and educationally beneficial child care. These enrollments and costs have been pretty stable over the years, which is certainly a desirable thing for forecasters. Factors affecting general enrollment levels include, at a minimum,
- mothers’ behavior in providing for their children (a constant), and using the subsidy program to help them (a process about which we know little);
- the capacity of private child care providers to supply care at the rates the State is willing to pay (reasonably predictable);
- the capacity of state staff to handle the applications and redeterminations that give families access to the program (again, reasonably predictable); and
- the State economy and supply of jobs for the mostly minority users of the program (reasonably predictable in the short term, in the long term much less certain).
The trouble with black swans is their color, of course—which is to say, their great rarity. One can study the history of any public program for some time without seeing them at all. When one does appear, there is little, usually, to be learned about where or when the next sighting will occur.
One of the most important origins of black swans is the invisible interaction of many of those fickle actors, clients, customers, people in short. Hence the uncertainty in the list above. Hence the strange crashes of the stock market such as the “flash crash” of May 6, 2010, when trillions of dollars of market value disappeared in a few seconds, only to reappear shortly thereafter. Hence the mystery of how people become aware of the operations of a public program they can use to protect their children. How do they react to rumors of changes in the program?
A child care black swan (well, more of a “gray” swan) appeared in Child Care Subsidy automated system records for March of 2011. For a reason never clearly explained, State payments to providers soared upward 30% that month compared to the average payment in months both before and after, even after normal adjustments for seasonal effects. The best guess: providers and staff, influenced by rumors of an approaching freeze on admission to the program, accelerated invoicing and reduced the backlog of applications ahead of the change.
Given that no one suffered much from this gray swan, and given that it was associated with a real policy change, we can be aware of a possible recurrence. But what can we do to predict other such outlier events? Very little, I’m afraid, for such is the nature of black, and even gray swans. It would seem that putting a great deal of effort into worrying about, and investigating possible black swan events in the public arena—unlike in the equities markets—may not be a fruitful use of taxpayers’ funds.