What is utility?

I had an email from a reader after last week’s post, thanking me for the Economics of Limited Editions posts but mentioned that they didn’t quite grasp the difference between value and utility. This is exactly the sort of feedback I appreciate as it helps me to realise when I haven’t explained something properly. This is particularly important with utility, as it is one of the most fundamental concepts in economics and one which is well worth understanding. So today, I thought we’d have a look at the idea and why it is so central to the field. 

In previous posts, I’ve explained that value is a measurement of the pleasure or satisfaction that we derive when we use a product. I think it’s quite an intuitive term: we immediately understand that some products are valuable and others are not, and that value is a subjective phenomenon. You might find a particular pen is quite satisfying (and therefore quite valuable) while I find it much less satisfying. On the other hand, utility isn’t quite as intuitive and I don’t think it conveys the subjectivity quite as well. It also seems more abstract, while value is something we have experienced. I tend to use value in general discussions and utility in technical discussions, but you should regard the two as synonymous. 

For most people, the idea of measuring value can seem a bit abstract. You might be used to distinguishing between something which is valuable and something which is highly valuable, but that might be the extent of it. Economists, on the other hand, go to great lengths to achieve a precise estimation of value: we try to convert our value for all products into a standard unit (e.g. dollars) so we can compare them, and understand someone’s preferences.

This can sometimes strike people as a bit offensive: it is outrageous that we could put a price on good health, a beautiful sunset, love, or even life itself. I can understand why someone might find that confronting, as once you have a price on something then it’s conceivable that someone might want to buy or sell that thing. But the goal of this isn’t to monetise but to understand our preferences and determine what things are most important to us in life. A beautiful sunset is certainly worth a great deal, but we might still place a greater value on good health. Putting a price on it isn’t a matter of selling it off, but understanding exactly what we prefer and by how much, preferences which exist independently of our ability to quantify them. 

Most of the time, the best technique for determining our utility is just to look at how much we are willing to pay for something. If we would derive $100 in utility from owning a nice FP then we would be willing to pay up to $100 to own it. This works well for products in which we have some experience but it doesn’t work quite as well for abstract goods or things which are beyond our experience. But, for our purposes, this is certainly the preferred method. It enables us to make comparisons of the value we derive from different products and determine which we prefer (where the value of one product is greater than another) or which we may be indifferent between (where the value of two products is equal). 

In economics, we take this a step further and assume that people are both rational and utility-maximising: that is, when faced with a decision, people will consider the options and their preferences opt for the one which yields the greatest utility — the most satisfying option. You can see that my model of the purchase decision is obviously based on this idea. People will compare various pens, think about how much value they get, and the price, and buy the one which yields the most bang for their buck. 

These assumptions give us a workable framework for understanding people’s decisions. As long as we know people’s preferences and the set of options available to them, we can compute the utility of each option and determine which choice they will make. We can understand if it was an easy decision (the utility of one choice obviously exceeded the others) or a difficult choice (the utility of two or more choices were close), and what were the relevant factors. It becomes a relatively straightforward exercise, one which we can use to understand past decisions or even to understand what someone will choose to do in the future. 

That understanding is quite valuable if you need to influence a particular decision. Maybe you want to convince a friend to try a particular restaurant, a manager to give you a pay rise, or a politician to support a particular policy: regardless of what the decision is, you need to start by understanding the utility of each choice for the decision-maker. Then you need to find a way to influence the utility of your desired choice: show your friend that the restaurant makes a dish he enjoys; help your boss to realise that the pay rise will make you more productive and increase his annual bonus; or help a politician to recognise that the policy will win him votes (or money). 

Utility maximisation is central to rational choice theory (RCT), the backbone of modern economic thought. It makes the claim that broad social phenomena is the result of millions of individual decisions, where each of us are making choices that maximise our own utility. So if you want to understand why society is a certain way, you need to consider things at an individual level. Alternatively, if you want to shift society in a particular direction, you need to think about how to influence all of those individual choices. 

A useful example of this might be the decline in cursive writing. Plausible explanations abound: it could be because of technology, changing social norms, a decision by schools not to teach it, or simply because it’s no longer valued in society. I don’t claim to know which answer is correct (if any of them are) but I am sure that any explanation needs to be rooted in rational, utility-maximising individual decisions. The idea of changing social norms is a bit abstract (even airy-fairy) for me to find plausible. But I am sympathetic to the idea that good handwriting once enabled clear communication and was valuable on that basis. Then technology enabled even better communication, and handwriting — a skill that takes a lot of time and effort to learn and maintain — became much less valuable, and schools ultimately responded by teaching subjects of relatively greater importance. This explanation makes sense at the individual level and it provides us with a good idea about what was the cause (technology) and what is an effect (choices by schools); if we want to effect some change in this phenomenon, this explanation suggests that we should focus on technology rather than petitioning our local schools. 

(It is striking how much historical argument and narrative is totally implausible when considered on this basis. If you want to understand why Cleopatra was a hussy, why ordinary Germans supported the Nazi party, or why soldiers enlist to fight in wars, always look to the explanation that considers individual utility, and ignore the broader social ‘themes’).

All of this comes together to form a behavioural science that is distinct from other approaches, such as neuroscience, psychology and sociology. Neuroscience sees the driving force of human behaviour as neurones in the brain: their structure, function, and the extent to which chemical imbalances influence their operation. Psychology sees the driving force as mental phenomena, such as the desires which emanate from the subconscious mind. Sociology sees social issues like power, class, and oppression as being the key drivers. To some extent, each of these fields are deterministic: they claim our choices and lives are principally determined by something or someone else. It is only economics that says the fundamental driver of human behaviour is our own rationality.

This focus on rationality has enabled economics to expand from the field of commerce and trade, into areas that have been traditionally studied by other disciplines. Gary Becker, who won the 1992 Nobel Prize in economics, was hugely influential in this. A sociologist by training, Becker adapted RCT and used it to analyse how individuals make decisions about whether to commit crime, to get married or have children, to invest in their education, or even racial prejudice. Operating on the assumption that the decisions were utility-maximising, he achieved insights which were not available with other forms of analysis and helped to identify solutions which would not have been recognised otherwise. Today, economics is present throughout academia, studying individual decisions as varied as architecture, culture, education, health, politics, religion, war. 

Of course, RCT is not without criticism. Right from the very beginning, its rigid assumptions were attacked: particularly assumptions that decision-makers have complete information (both about our available choices and our own preferences), that we are always and ruthlessly utility-maximising, and that our preferences do not change over time. I think most economists would find that these criticisms are entirely correct: those assumptions are completely unrealistic, and not at all representative of how people make decisions. The real question is whether that matters. 

Milton Friedman (a giant in economics) made a compelling argument in the early 1950s that such criticism was irrelevant. As long as a theory is able to make accurate predictions about a wide range of phenomena, he said that we should not care about the realism of the theory’s assumptions. This was, and remains, a controversial view but it is one implicitly accepted by many economists (and those who rely on economic analysis). 

My own model is a good case in point: while some of us might actually think about value, price, and surplus, I suspect that most of us make decisions on a less rational basis: we might strive for a good deal but we’re not doing much in the way of analysis to determine if we’re achieving an optimal result. And yet, it doesn’t really matter if the model accurately represents our thinking. At the end of the day, we do basically end up with the pen which offers us the greatest surplus, and the model does provide some useful insights into the important factors in the decision. 

So it’s entirely possible unrealistic assumptions genuinely don’t matter, but some economists have nonetheless responded by developing theories with  more realistic assumptions. The best known is behavioural economics, which largely dispenses with the rationality assumption and instead runs experiments to empirically understand decision-making. It has had a lot of publicity but, at least in my opinion, suffers from the same problem as sociology: lots of rich, highly nuanced case studies but little in the way of conceptual frameworks or general theories that we can usefully apply. 

Another response was my own field, New Institutional Economics (NIE), which has carved out something of a middle ground. It claims we may strive for optimal decisions but our access to information is limited and our rationality has bounds (or limits); this means that we often fall short of the ideal. While this approach has led to some fruitful insights, particularly into dynamic phenomena (how things develop and change in the long run), the output has been quite limited in comparison to RCT.

Despite the unrealism of its assumptions, RCT has proven to be remarkably durable and useful. Academics, analysts, and policymakers routinely use RCT (particularly it’s progeny, the cost-benefit analysis) to understand, to predict, and to make optimal decisions. In the centre of this theory is utility, which means individual satisfaction and pleasure is right at the very heart of economic decisions. It’s not a story that gets told very often, but it’s an important thing to keep in mind the next time you see someone arguing that we are people, not numbers. 

So this has been a bit more technical and econ-focussed than most of my posts, but I hope that it sheds quite a bit of light on the idea of utility and my model. And, as always, thank you for reading!