It's not because my cheery disposition causes me to celebrate small gains and shrug off big losses: as my acquaintances will testify, it's not difficult to distinguish between me and a ray of sunshine. Instead, it's because I do what millions of others do. We have separate mental accounts, so we think of £50 in one account as being different from £50 in another.
Not only does this contradict basic common sense - £50 is £50 - it can also lead us into expensive mistakes, as the University of Chicago's Richard Thaler has shown. For example, it can cause us to hold onto poorly performing shares if we have different mental accounts for paper losses and for realised ones - because the greater pain of realising a loss will stop us selling. This can be costly because there is often momentum in shares; our own negative momentum portfolio (which comprises the 20 worst-performing large stocks) has underperformed the FTSE 350 by over forty percentage points in the last three years. This means that holding onto losing shares causes us to lose even more.
It seems therefore that all my talk of cognitive biases and investment errors has done me no good. I'm as prone to irrationality as everyone else. This fits a long-standing pattern. Even the smartest people can be very silly sometimes, as the pronouncements of James Watson and Richard Dawkins regularly remind us. If they can be daft, so can I.
Or am I? I'd defend my particular mental accounting on two grounds.
First, a stock market gain or loss is not the same as a premium bond win. The latter is for keeps, the former not.
What I mean is that there are two common ways in which shares can rise: either because appetite for risk increases so that investors require lower future expected returns to compensate them for holding shares; or because investors become over-optimistic about growth. In either of these cases, rising share prices today mean lower returns - or even negative ones - in future. This means that day-to-day fluctuations in share prices tell me nothing about what really matters to me, which is my likely income when I retire. By contrast, a £50 premium bond win tells me I'm £50 better off.
Secondly, the rule 'spend premium bond wins but not stock market gains' is a reasonable one. It gives me small occasional treats while preventing me from being tempted to spend too much.
Of course, this rule looks odd from the point of view of rational economic man who maximises utility subject to an intertemporal budget constraint. But such a man is a fiction, who is not afflicted by weakness of will. Those of us who are real human beings are so afflicted - and my rule of thumb is a Ulysses pact which protects me from that weakness.
What's at stake here is not just the trivial matter of how I approach my investments. It's about the very nature of rationality itself.
First, whatever you think of the merits of these two arguments, they are rationalisations; they come after me celebrating my premium bond prize and are not, I suspect, the motive for it. As David Hume famously said, "reason is, and ought to be, only the slave of the passions". Luckily for me, in this - perhaps isolated - example, my passions serve me well.
Secondly, this raises a question: are rules of thumb which deviate from economists' conception of fully rational behaviour justified or not? On the one hand, research by Nobel laureate Daniel Kahneman says yes: the entire field of behavioural finance is founded upon this. On the other hand, Gerd Gigerenzer at the Max Planck Institute argues that some simple heuristics can actually improve our decision-making.
Who's right? It depends on the context and the rule. Which is my point. We should check our instincts and our rules of thumb and ask: are these consistent with my objectives and the empirical evidence? Sometimes - as with our tendency to hold onto losing stocks - the answer is: no. But, at other times, perhaps they are.
