To what extent is the Scottish referendum about the benefits of independence and to what extent is it about people's attitudes to risk?
To see my question, think of independence as an investment with a risky payoff - which is what it is.
There are two reasons why one might reject such an investment. One, trivially, is simply that you expect a negative payoff. The other is that you expect a positive payoff but think the risks around it are too high.
It is perfectly possible for two people to agree upon payoffs and risks and yet one would accept the proposal and the other reject it because one is more risk-tolerant than the other. This is commonplace in asset allocation. Two investors might agree upon expected returns and volatility and yet one might hold more equities and fewer safe assets than the other simply because of differences in tastes.
Mightn't the same be true for many people for Scottish independence? Two Scots might agree on the payoffs and risks to independence but one might vote yes and the other no because of differences in risk tolerance.The statement "Independence probably would be a good thing, but I don't want to risk it" seems tenable to me.
There's an analogy here with the Duhem-Quine thesis. Just as experiments are usually tests of joint hypotheses, so too are referenda; they test attitudes to risk as well as expected payoffs.
One fact lends credence to this. It's that women seem more inclined to vote no than men. For example, the latest YouGov poll shows men 51-44 in favour of independence but women 39-55 against. I don't see why men and women should have different knowledge about the effects of independence. But I do know that there's evidence that women tend to be more (pdf) risk-averse than men. And this would tip more into the no camp*.
Now, this isn't decisive. There's less evidence for gender differences in uncertainty-aversion - and this (unknown unknowns) might be more relevant than risk-aversion (known unknowns) to the vote.
Nevertheless, it's possible that in a tight contest the referendum might be decided not just by the merits or not of independence, but by attitudes to risk. A no vote might tell us not that independence is a bad idea, but that it's a good one but some Scots are too risk-averse to chance it.
If I'm right, there are some implications.
One is that, from a nationalist point of view, autumn is the wrong time for a referendum. We know from the stock market that risk appetite is seasonal; it's high in the spring and low in autumn - which is why the "sell in May, buy on Halloween" rule usually works so well. Because the Nats want a high appetite for risk, they should have gone for a spring vote.
However, all is not lost for the Nats. There's also some evidence that risk appetite is affected by the weather - it's higher in sunny days than wet ones. If the Indian summer lasts into next week, it could tip some into the yes camp. Not many, granted - but in a tight vote, even small influences count.
Secondly, if Scotland does vote no, the Nationalist cause won't die. Everyone has immunizing strategies which protect their cherished beliefs from challenge. Nationalists could claim not that Scots believe independence to be a bad thing, but that some were too risk-averse to go for its benefits. It'll be very difficult to refute this.
* Whether this is because of nature or nurture is moot but irrelevant for my purposes.