Archive for August, 2015

The fairest of them all

For all our notoriety about being unemotional, unsocial, and unfeeling, economists have a preoccupation with issues of human happiness, at least from a theoretical point of view. One of the things we obsess over is something that we haven’t quite solved, which is the concept of fairness.

Recently, issues of fairness came up in the form of the sport that I love, where the debate is over prize money: should men and women get equal prize purses when there are less women competing? Since this is exactly the kind of thing I do for a living, this kind of philosophical question is going to drag me into dangerously nerdy territory.

The problem of the pie

Distributional issues involve trying to split two different ‘pies’: either I’m trying to divide up the costs of things between various people, or I’m trying to divide up wealth.  It is essentially the same thing; let’s focus on the cost allocation question in the simple example of postal services, good old fashioned letters you write with your hands. It becomes complicated when:

  • There are different kinds of costs. Some costs exist even if nobody sent any letters and are fixed (the rent on the storage buildings), and some are costs based on the sender/recipient (the gas for the trucks that need to drive out to the postal locations) and some are based on the number of letters (the number of workers at the postal processing centre).
  • Every letter costs something different. Some people are sending letters from London, to London, and others are sending them from the Shetlands to Fattiehead (I looked that one up, it DOES EXIST, lol – it was probably the least offensive English town name on the list of weird English town names). The cost for the storage buildings is the same for both letters but the truck that drives from the Shetlands to Fattiehead, the cost is much greater.

The question appears when you decide how to set the prices, because the costs need to be recovered from people who are sending letters. There are two extremes:

  • Everyone for themselves. One way to do it is if everyone pays the amount to send the letter that it costs the post office to send it.  That means services provided in central London are dirt cheap, and ones in the Shetlands are very expensive.  In a related example, if the cost of goods is reflected in the cost of shipping it wherever it needs to go, this explains why costs of food are several times higher in remote parts of Canada, where the price reflects the actual cost of getting food to those remote locations.
  • Smear the cost equally. Another solution is to charge everyone the same thing, regardless of if they are sending a letter from the Shetlands or central London. This means that letter senders in London are subsidizing service in other parts of the UK by covering more than their share of costs.
  • Some kind of compromise. Maybe the charges are done by zones; the same base charge for everyone, but ones that are north of central England pay a fixed higher charge. There are a variety of ways to design a system of postage stamp prices which would be somewhere between totally uniform prices and totally unique prices.

The problem has problems

There are major questions which need to be answered in designing a good way to cut the pie.

What about the different needs of different people? London has a significnatly higher level of income per capital than more remote parts of the country. Maybe access to electricity, internet, and postal services is poorer outside of Southeast England and these services are needed in order to encourage development. Should a concern for need mean that those who have lower costs shoulder a larger share of the burden?

What’s going cause the least interference? Economists are super concerned about ‘distortion’, and try to minimize the ways in which any cost or tax policy deviates from what would naturally happen in a market. This means that there is a certain pressure to maintain a cost structure that reflects what costs users of the system are actually incurring. According to the Ramsey rule for optimal commodity taxes, the most efficient way to design a tax system would be to impose the highest level of tax on things that people really need (milk and bread), and the lowest on things that are optional (luxury vehicles). (This is sometimes considered super unfair and will be discussed later). If the postage cost is the same across the country, then this may encourage more people in remote places to use post (increasing the cost to the system) and cause those in low cost areas to look for other ways of getting letters delivered (lowering the amount of subsidizing users). The gap in available funds to cover the cost is caused by the way the pricing encourages people to behave in ways they otherwise wouldn’t.

What’s efficient? Efficiency is considered in the sense of who would make the best use of a reduction in postage cost. Maybe the businesses in London are larger and more important for overall productivity than remote firms. Making the productive firms subsidize the cost means they have less available funds for other investments.  Also, maybe the postage system doesn’t *want* to encourage people to set up shops in the Shetlands, since that would be expensive, and keeping the price of business down in London would encourage more people to locate their businesses in the same place. Efficiency and distortion are often related concepts.

What is politically palatable? UK postal provider Royal Mail actually does have a Universal Service Obligation to serve everyone and charge an equal rate. This is done for both political and fairness reasons, which are often intertwined.  It is hard for a policy to publicly justify charging or taxing different groups differently, and society may have an innate concern for fairness and the protection of minority/vulnerable groups. This may mean that society is willing to forego a certain amount of efficiency or undergo a level of distortion to achieve a fairness aim, although the concepts of fairness, efficiency, and distortion don’t necessarily have to be at odds. For example, even though Ramsey would suggest the best way to tax goods is to tax basic necessities higher than optional luxury goods, it may be hard to sell such a policy to a group of poverty-sensitive voters.

What’s easy? In another example, electricity grids, and the generators that flow electricity through these grids, are often owned by two or more different sets of people. In order to determine how much to charge a generator for the use of its grid, the grid operator would need to know where power is being introduced/drawn out of the grid, in what levels, at what locations, the cost of the grid, and all of this would need to be measured in real time. The different cost that one generator imposes depends on every other user. Also, this would need to be *predicted*, as they are probably paying a price now and we don’t know where the power is actually going. The measurement of this is no small feat of mathematics and engineering, and it takes more money and effort to figure out how much to charge users than simply adding up the total and dividing by everyone. At the end of the day, after the debates about fairness, efficiency, etc, the real deciding factor is often which way is the cheapest way to figure out how to cut the pie.

What to actually do?

Out of the three ways to divide costs (everyone for themselves, smear the costs equally, or compromise), one seems to be more immune to public scrutiny than the other two, which is to just promise equal costs.  Is this absolutely the fairest, most efficient system based on the different needs of the people? Probably not in all cases, but it does not open the pie-cutting process itself up to scrutiny. For example, if everyone were to pay a different cost of postage based on a different cost of delivery, trying to identify everyone’s costs will be difficult; what about the always changing price of transport? What happens if two neighbours in a remote town decide to send letters on the same day? Does the price get halved? The pricing structure would need some kind of a system for justifying differences in pricing in order to avoid accusations of arbitrariness and unfairness.

Out of all the ways to allocate costs, there are winners and there are losers, and if the losers know they are losing, they may cry foul. While its  potentially better to think about what’s an efficient, non-distortive way to split costs, sometimes the easiest thing to do is to give an equal share to everyone unless there’s an extremely compelling reason to do something differently. It better be damn compelling. All the competing factors play may have legitimate arguments in favour of one way of cost splitting over another, but we live in a world where political palatability and doing something convenient/feasible tends to win out over other kinds of reasoning. Want to have a sports tournament? Pay gender divisions equal prizes. Don’t want to do that? Be prepared to show everyone the math you have done. Don’t have any math? See the first suggestion.

The biggest lesson you can probably draw from this is that economists spend a lot of time trying not to upset people that we allegedly have no feelings about.


Read Full Post »