Letter from Germany: The return of the watering can and some Oktoberfest economics

Have you ever wondered how to construct the optimal incentive contract for staff serving in the chaos of a beer festival? Ray Rees, at the University of Munich, reveals exactly how it’s done at the city’s famous Oktoberfest.

In last year’s letter, I told the story of how the academics on the Commission set up to allocate €1.9bn over five years under the German government’s ‘Excellence Initiative’, caused alarm and consternation among the politicians representing regional interests by choosing only three universities — two in Munich and one in Karlsruhe — as suitable to be ranked as excellent. So the question was whether, in the second round held this year, the politicians would strike back and restore the ‘watering can principle’, under which funds are spread around the regions more or less equally, and independently of differences in academic merit. There have been some claims that this was the case. A further six universities were awarded the accolade, including the grand old institutions at Göttingen and Heidelberg, while many more universities were given funding for specific graduate schools or ‘excellence clusters’ — large inter-disciplinary research groups. The three universities that were successful last year received virtually nothing in this round, and indeed had their funding cut back marginally. So, was this the revenge of the politicians, the triumph of regional interests over excellence?

The criticism that the watering can principle had returned seemed to stem largely from disgruntled applicants, particularly in the arts and humanities, who had had their projects rejected. The head of the country’s Scientific Council, the equivalent of the ESRC, and himself a professor of German Literature at the University of Munich, denied that this was the case, and said that the academics on the Commission were fully in support of the proposals. The only criterion had been academic merit. The awards tend to bear this out. There was again a strong bias towards the South and West of the country, with the big civic universities in important cities such as Frankfurt, Cologne, Hamburg, Bonn and Kiel being denied — absolutely realistically — excellence status. In Berlin, the smaller Free University was chosen over its rival the Humboldt University, and no other universities in eastern Germany were anointed. Had one been so, that would have been a sure sign of the triumph of regional policy.

The dust has settled, and the big question now is whether this major departure from the previous doctrine, maintaining that all German universities are equal and must be treated equally, will succeed in re-establishing the standing of German universities in the scientific world. One apparently good sign: two German scientists, working in Germany, received Nobel prizes in physics and chemistry respectively, shortly before the excellence awards were announced. But in a sense this underlines the problem. Both worked at specialised research institutes rather than at universities. The post-war decision to downgrade the research status of the universities and overload them with students, concentrating research on a small number of non-teaching, well-funded research institutes, was a disaster for the universities and, last year’s prizes notwithstanding, has probably damaged Germany’s overall research performance. The Excellence Initiative is a partial reversal of that policy. In 2011, when the whole exercise will be repeated, we should have at least the preliminary indications of success or failure.

Oktoberfest Economics1
For many people living in and around Munich, the annual Oktoberfest, which despite its name is held mainly in September, with just a small overlap with October, is the high point of the year. Take a very large fairground, with the most technologically advanced and stomach-churning rides in the world, add to it 14 large beer tents, ranging in size from 3000 to 8000 places inside and 2-4,000 in the beer gardens outside, with then a myriad of smaller tents for more specialised eating and drinking opportunities and all kinds of stalls and sideshows, and you have the basic infrastructure. In the 18 days of its existence in 2007, it is estimated to have generated €955mn in revenue, with €455mn earned at the Oktoberfest itself, €200mn being spent in Munich’s stores, and €300mn in hotels. For the rest of the year, this large area of prime land in the centre of Munich, called by Münchners the Wies’n, is just a large car park, and so it has to earn its annual opportunity cost rate of return in just those 18 days. At the same time, it is very doubtful if any city politician would be prepared to face the outcry that would result from a proposal to hand the Wies’n to property developers, and so, given this constraint, its actual opportunity costs are hard to estimate. Despite the fact that it is a major tourist attraction, around 70 per cent of its visitors still come from Bavaria, and 60 per cent from Munich.

The tents open at around 10 am. and fill up steadily through the day, often becoming completely full by 5 pm., especially if it is raining, when the outside beer gardens aren’t much in demand. One of the favourite postcard scenes from the Oktoberfest is of a buxom, rosy-cheeked waitress dressed in traditional costume and carrying an amazing number of foaming one-litre mugs of beer — often around ten — and still managing to smile. One of the most interesting things about the Oktoberfest, from an economist’s point of view, is the form of the contractual relationship she has with the owner/operator of the tent in which she works. She is employed just for the duration of the Oktoberfest, and has probably the most high-powered incentive contract you will find anywhere. She has no fixed wage, but receives 9 per cent of the revenue on the beer she collects from the serving points and brings to the table, as well as tips of course. Thus she has the incentive to maximise sales revenue as well as give good service.

This is not a trivial thing. She works more than a 12-hour day, and the fetching and carrying is very hard work, particularly when the tent becomes full and turns into an enormous disco, with people dancing on the benches and waving their beer mugs around. This usually happens from around 6 each evening. It is easy to see that a fixed wage unrelated to revenue would have very poor incentive properties. A closely complementary task is actually to collect the money. People tend to wander away from their tables from time to time, not least because of the inevitable consequences of pouring large volumes of beer down their throats, and the usual method of collecting payment in German pubs and restaurants, when the client is finished and ready to leave, would be a very risky business here. So everything is paid for immediately as it is delivered.

The revenue-sharing form of contract means that the presumably risk-averse waitress receives no insurance, in the form of a fixed payment, from the tent owner, who is far richer and therefore likely to be less risk-averse, if not risk-neutral. This leads one to ask whether there may not be an alternative way of providing incentives. A noticeable feature of one of the large beer tents is the variation in waitress effort and rewards there must be across locations. Tables differ in their distance from the serving points and therefore in the work involved. Some areas fill up earlier than others, for example those closest to the bandstands, and therefore generate more revenue. In particular, there are areas of the tent reserved in ‘boxes’, which are booked by companies and groups, and these generate more tips than other areas, as well as higher-value sales. So one could imagine a tent operator running an incentive system in the form of a tournament, in which high effort, as indicated by revenue, would be rewarded by allocating waitresses to the best groups of tables. In fact however this does not take place. Waitresses as a rule serve the same set of four tables in the same tent year after year, and moves between areas are infrequent.

The reason for this appears to be the importance of teamwork, as well as long-term customer relationships. Usually, a team of four waitresses work their set of four adjacent groups of four tables collectively, keeping the same team over a number of years. They share the fetching and carrying, which smoothes the burden of work and allows more regular individual rest periods, and they allocate arriving customers across tables, which effectively pools revenues. If one member of the team decides not to work in the next year, she tells the others and a replacement is found who is usually a relative or friend of the remaining members. Moreover, customers form attachments to particular groups of waitresses and return year after year. A tournament system would disrupt these kinds of relationships, while of course they provide self-made insurance and incentive mechanisms.

The price of an Oktoberfest Mass, the litre mug of beer which is the standard unit of consumption, is regulated, and is set each year by a committee consisting of representatives of the tent operators and of the City Council. Its announcement is always followed by a storm of protest — the increases are usually above the rate of inflation. For waitresses, the crucial numbers are those after the decimal point. Apparently, customers almost always round up to the nearest Euro, so a price of €7.80 is bad news, €8.20 good. For a waitress serving around 1000 Mass per Oktoberfest, this would make a difference of about €600 in tips. On the other hand, since waitresses typically work over a long run of years, it all averages out in the end. There is no data enabling us to test whether the rate at which waitresses quit is sensitive to the price of a Mass.

Love it or hate it, the Oktoberfest is a fascinating economic enterprise. It certainly gives economists plenty to talk about over our beer.


1. I am grateful to my student Andreas Moser for his research on the information on which this discussion is based.

Page Options