Donate to Radiohead

As I’m sure you’ve heard by now, Radiohead have decided to give away their new album online in exchange for a voluntary donation. When one buys the album you choose how much you want to pay for it, with no lower or upper limit on how much you can pay. the idea of paying for something that you can get for free is one that’s puzzled economists for some time. The classic experiment in the field is the Dictator game. In this game one person has an endowment of money and they choose how to split it between them and another person. Despite having no pressure to do so, many people give money to the other person in the experiment. There have been a truckload of different experimental designs employed to investigate this phenomenon.

The key difficulty is that people have the opportunity to donate money to strangers in everyday life but few tend to. Why is it that they then give money away when faced with this experiment? Most explanations focus on the contrived nature of the situation: with a researcher standing there observing your donation one worries about being thought of as selfish if one keeps all the money. Indeed, a large scale anonymous experiment with no researcher present showed very little evidence of altruistic donations.

So what’s my pick for the Radiohead sales? Well, the donation is unobserved and anonymous so the evidence suggests that casual listeners are unlikely to pay for it. However, publicising how much you donated for the album could be a credible signal of your dedication to the band. I’d expect fans of the band to give fairly generously and make their donation known to others. I see also that to get the album you have to enter your credit card details, however much you intend to pay. This seems like a good way of reducing the marginal transaction costs of giving: if you have to enter your details anyway then it’s easy to give a couple of dollars at the same time.

Orchardists and Labour productivity

NZPA released a story about an Orchardist that tried to get out of paying some seasonal workers for a public holiday.  All well and good, the contractual obligations of an employer is a topic that is out of my league.  However, the final line of the article got my interest.

“When employers treat their workers well by paying their entitlements, their workforce is likely to be much more productive”.  This is the claim of the Department of Labour.  Now I think by itself this is a fair claim.  In the apple picking industry it can be difficult to quantify the amount of effort a worker is putting into picking.  The output of the worker depends on both the effort they put in, and the density of the fruit in the area they are picking (I did a little blueberry picking back in the day 😉 ).  It is also impractical to constantly supervise workers (as they tend to work over a largish area).  As a result of these factors, efficiency wages can increase effort and thereby increase the workers productivity.

However,  the context that DOL made this statement in wasn’t purely descriptive.  They were trying to tell employers that they should be more generous with their wages, as they should want higher productivity.  In this sense I disagree with them.  The employer realises that the productivity of their workers depends on the way they treat their workers.  If they choose to pay their employees at a low rate, it is because the expected benefit from paying them more is less than the cost of paying them more.  Now in the case of the apple orchard this was illegal.  However, the employer obviously felt that the probability of getting caught was sufficiently low that the cost of paying his employees (which includes the productivity enhancement, and the loss from getting caught times the probability of being caught) was less than the benefit from paying.

If this is DOLs line on productivity, they are treating employers like they are stupid.  They believe that they understand the relationship between employers and employees better than the employers and employees themselves.  While I do not have a problem with the idea that higher wages leads to greater productivity, I do have a problem with the idea that firms are not doing what is in their own best interest.  If this is how the DOL feels, they need to realise that employers goal is not to maximise the productivity of their workforce, it is to maximise the profitability of their business.

Some people may feel that is would be a good idea to intervene and force firms to pay higher wages and increase labour productivity.  If we did this output could rise or fall (depending on the relative effect on productivity), the amount of labour hired would fall (as you would need less labour to produce the same quantity of output), and the effect on prices would depend on the change in output (as we are moving along the demand curve).  Ultimately, we assumed that the firm would not want to do this unless the benefit was greater than the cost, if this is the case output from each firm would fall and prices would rise.  In the apple industry we face a world price, and for the consumer, the loss of output would be made up by imported apples.  However, the leftward shift in the domestic supply curve implies that producer surplus would fall.  As consumer surplus hasn’t changed, total surplus from the industry would fall.

So by forcing firms to set higher wages, to force a higher level of productivity than they would have chosen in equilibrium, we get greater unemployment and lower profit in our apple industry.  Not an outcome that people on either side of the political spectrum would be particularly happy with.

Note:  I am only talking about setting higher wages to receive higher productivity and how that influences efficiency.  I am not talking about the equity reasons for higher wages, and I’m certainly not talking about the minimum wage.  You can talk about that stuff if you have a point you want to raise, but if anyone gets all ideological and angry about it I’m going to be very mean to you!

Counter-signalling

Yesterday’s post on dress codes and signalling drew a comment pointing to Tyler Cowen’s reference to ‘counter-signalling’. TC refers to Steve Jobs dressing down as a counter-signal because he doesn’t need to dress up to show his seniority and importance. It’s a bit of a misleading term because a counter-signal isn’t really a signal at all: it doesn’t convey any information since it is not costly to fake. Anybody can dress down but for many people it’s not conducive to good career prospects. Essentially, Steve Jobs doesn’t need to signal his enthusiasm to his superiors because he has none, so he doesn’t signal at all. The fact that he doesn’t need to signal isn’t itself a signal because anybody can do nothing.

The original paper that this is drawn from goes a little further than pointing out that signalling is unnecessary for some people. It says that when a signal is easy to make – such as wearing a suit – then it is detrimental to high quality people to make the signal. The information conveyed by a signal that’s easy to make is that one is, at worst, mediocre. When you are, in fact, a high quality candidate then it can be damaging to you to be seen as feeling the need to prove yourself to be mediocre. Your optimal strategy is to avoid making the signal at all and look for a better signal of high quality. Thus, the people who end up making a signal are the worst of the people who are able to make that signal.

I guess the implication of this for dress-codes is that the people in nice suits are the lowest level employees who can afford to buy that quality of suit. Once you’re at a level where you can afford better suits you should cease wearing the lower quality suit at all. When you’re at the top and your subordinates wear top-drawer kit then you’re best off avoiding suits altogether and making no signal via your clothing.

Dress codes in the office

I have always wondered why corporate offices require a dress code for employees, even when the employees never see the outside world. Tyler Cowen blogs about corporate dress and makes a couple of interesting points, but it doesn’t really address my particular curiosity. He’s inspired by a reader who points out not many people are as comfortable in business attire as they are in casual clothes. If a worker is most productive when they are most comfortable then why would you force them to wear a suit?

A common explanation of smart clothing is signaling to overcome adverse selection. People choose to bear the cost of wearing uncomfortable but smart clothes to show how dedicated they are to their job and to gaining a promotion. However, this seems like an argument against dress codes: if people can choose what to wear then it’s easier for the eager employees to signal their dedication to The Company through their choice of clothing. As Tyler mentions, it may even be worthwhile for companies to impose a maximum dress code in order to eliminate the signaling costs that allowing suits imposes upon employees.

Ultimately, I think the reason for dress codes is shrouded in the mists of corporate managerial lore, of which I know nothing. I can’t think of a good reason for it in standard micro theory, and since Tyler Cowen can’t either I’m in good company there. If anybody has a good explanation for this convention I’d love to hear it.

The marginal revenue of movies

I was reading the Dilbert blog the other day, and Scott Adams was trying to talk about a Dilbert movie. Before starting to ask the blogging community whether they thought it would be a hit, he linked to this article, and asked why movie producers keep making so many R rated films, when G rated films gave larger returns. He puts it down to the directors incentives not being aligned with those of the investors. So instead of making a film that will maximise profit, directors make films that maximise the chance of winning awards. This is all very good, and I’m sure that there is an agency problem in film making. However, the agency problem will be ex-post the decision on what type of film to make. It seems to me that if investors felt they would make the most money of G-rated films, they would make contracts with directors to force them to make G-rated film, thereby solving (most) of the agency problem.

Scott Adams has an extremely good understanding of economics (he did do a degree in it), however, in this case I feel that he has confused average and marginal revenue. Now the article states that the average revenue from a G-rated film is greater than the average revenue from a R-rated film. However, when an investor puts money into a film being made, they are interested in the marginal revenue of that film type that is available from the market. The investor is putting money into a project that creates 1 more of that type of film in the market place. In that case the marginal revenue of a film type is what the investor is after.

Now it is possible that the average revenue of R-rated films may be lower, while the marginal revenue of an R-rated film is the same as a G-rated film. All we need for this to happen is the initial returns from the first G-rated film on the market to be higher than the R-rated film, but for marginal revenue to then fall much faster for the G-rated film type than the R-rated film type. I think this is a good description of demand for movies, so people can see them on their TVs in their homes or gardens with the use of accessories as an outdoor tv wall mount for this purpose. Families have a high value for taking their kids along to the first great animated film of the year, however once we get to the 3rd animated film for the month the family will stop going. However, R-rated films tend to get a specific crowd that is interested in that type of film. This crowd is willing to pay to see more movies, but may not be as large as the crowd that goes to the first animated film. Furthermore if there is more range among R-rated films than G-rated films, the set of products we are discussing is catering to a differentiated market. If this is the case, the release of an R-rated film would have less of an impact on the marginal revenue of another R-rated film than would be the case with G-rated films.

However, there are a number of other explanations for the difference in average revenue:

  1. R-rated films are less prone to failure (so the marginal profit can be lower)
  2. R-rated films are cheaper to make (so although marginal revenue is lower, marginal profits are equal)
  3. There is some bias among investors towards R-rated films
  4. There are too few directors and too many investors, and so directors they have some market power. As a result, directors can change the composition of movies

Personally, I think all these things come into play in some way. However, I’m confident that movie investors are often trying to maximise their profit, just like people who invest in companies or houses.

Airlines and competition

I’ve been thinking about the fact that AirNZ is going to shut down Freedom Air in March 2008. With Freedom Air, Air NZ was able to serve the budget end of the market and the higher quality end by selling a differentiated product. However, the company could have simply offered different services in different compartments of the plane, it seems a touch over the top to create a whole separate brand just to get the advantage of price discrimination (at least in this case).

The true purpose of Freedom air was to prevent competition. By paying a whole lot of money to run a cut price airline, Air NZ was able to make it uneconomical for other potential entrants from coming into the market, as their marginal cost was greater than the price they could set when competing with Freedom. The investment in Freedom air acted as a form of commitment. As the investment in capital, goodwill etc for Freedom air was costly to reverse, Air NZ could credibly commit to fighting a new competitor on the Trans-Tasman route through the low prices Freedom charged, rather than just allowing them entry.

However, Air NZ has dropped Freedom right when Virgin was getting in on the act. I expect this is because the costly commitment to fight the new competitor was not sufficient to prevent the new competitor entering. As a result, Air NZ has decided to dump Freedom Air and just accept that there is a new competitor. I guess this is fair enough, as Virgin has some very deep pockets, and if Air NZ decided to fight them they may well be on the losing end. As a result, Freedom Air was a useful mechanism to reduce small scale competition in the market place, but it was not effective at preventing the arrival of one of the big boys.

In the future, I’m sure the case of Freedom Air will be a useful case study in how an incumbent can use costly commitment to prevent the entry of a new competitor. How do you think the commitment game functions in this case?