I remember a commercial for some kind of diet program where that was the tagline. A disembodied hand kept enticing this poor guy with delicious looking food and then taking it away because it was unhealthy and then the voiceover came in with that line and I thought that was so tragic that everything that tastes good had to be bad for you. Like what kind of cruel joke is that?
And it makes no sense from a biological point of view. I should want to eat what’s good for me so that I do eat what’s good for me and avoid what’s bad for me. That’s Mother Nature’s optimal incentive scheme. And once we have evolved to the point that we can think and understand that principle we should be able to infer that whatever tastes good must in fact be good for us. But it’s not!
At the margin it’s not. Indeed the right statement is “If it tastes good then you surely have already had too much of it to the point that any more of it is bad for you.” Because the basic elements in food that we love, namely sugar, salt, and fat, are all not just good for us but pretty much essential for survival. And so of course we are programmed to like those things enough that we are incentivized to consume enough of them to survive.
But the decision whether to eat something is based on costs as well as benefits. Nature programmed our tastes so that we internalize the benefits but it’s up to us to figure out the costs: how abundant is it, how hard is it to acquire, and when it’s sitting there before us how likely is it that we will have a chance to eat it again in the near future. Then we need to weigh the costs and benefits and eat up to the quantity where marginal costs equal marginal benefits.
It’s interesting that Nature put a little price theory to use when she worked all this out. A price is a linear incentive scheme. Every additional unit you buy costs you the same price as the last one. Your taste for food is like a linear subsidy, every unit tastes about as good as the last, at least up to a point. When you face linear incentives like that you consume up to the point where your personal, idiosyncratic marginal cost equals the given marginal benefit. If a planner (like your Mother Nature) wants to get you to equate marginal cost and marginal benefit, a (negative) price is a crude incentive scheme because the true marginal benefits might be varying with quantity but the subsidy makes you act as if its constant.
But that’s ok when the price is set right. The planner just sets the subsidy equal to the marginal benefit at the optimal quantity. Then when you choose that quantity you will in fact be equating marginal cost to the true marginal benefit. That’s a basic pillar of price theory.
So Nature assumed she knew pretty well what the optimal quantity of sugar, salt, and fat are and gave us a taste for those elements that was commensurate with the true marginal benefit at that optimal point. And its pretty much a linear incentive scheme at least in a large neighborhood of the target quantity. Sugar, salt and fat don’t seem to diminish in appeal until we have had quite a lot of it.
The problem is that the optimal quantity depends on both the value function and the cost function. Now the value function, i.e. the health benefits of various consumption levels is probably the same as it has always been. But the cost function has changed a lot. Nature was never expecting Mountain Dew, Potato Chips and Ice Cream. The reduction in marginal cost means that the optimal quantity is higher, but how much higher? That depends on how the shape of the value function at higher quantities. The old linear incentive scheme contains no information at all about that.
But one thing is for sure. If the true marginal benefit is declining, then at higher quantities the linear incentive scheme built into our taste buds overstates the marginal benefit. So when we equate the new marginal cost to the linear price we are doing what is privately optimal for us but what is certainly too high compared to Nature’s optimum. If it tastes good its bad for us we because we have already had too much.