Discussion Paper

No. 2008-28 | September 24, 2008
Perfecting Imperfect Competition


This paper addresses the reduction of market failure under imperfect competition. It proposes a tax-scheme that provides firms with an incentive to forgo their market power: Firms optimize after‐tax profits. Now simply consider a firm’s gross profit margin the unique tax‐rate it is charged on absolute profits. In theory the firm’s tax‐rate would be the mark‐up over marginal costs, the firm’s Lerner index. As a result every firm determines its own tax‐rate by setting its price and incurring costs. This creates a new trade off for firms between a low tax‐burden and the exercising of market power. Welfare for society increases since firms with market power choose a lower price and produce a quantity closer or equal to social optimum; at the original monopolistic price‐level they can increase their profits by lowering their tax‐burden. Essentially the tax‐condition does not seem to distort profit incentives or markets; under perfect competition the tax‐rate would be zero. Thus, it is clear that the tax only takes effect when markets work inefficiently and its countervailing nature subsequently helps to remedy inefficiencies of imperfectly competitive markets.

Data Set

JEL Classification:

D00, D21, D40, H21, H25, H26, P11


  • Downloads: 3055


Cite As

Goetz Seißer (2008). Perfecting Imperfect Competition. Economics Discussion Papers, No 2008-28, Kiel Institute for the World Economy. http://www.economics-ejournal.org/economics/discussionpapers/2008-28

Comments and Questions

Anonymous - Discussion Paper 2008-28
November 03, 2008 - 13:09

The comments are attached herewith.

Goetz Seißer - Response
November 24, 2008 - 01:41

To point 1: I assume that most readers read the abstract where the basic intuition is stated. I didn’t want to be repetitive. So I used the introduction to quickly put the condition into perspective with competition and regulation in order to move on and explain the workings of it ...[more]

... more elaborately in the main section. I have decided to leave it that way.

Point 2: I agree that it may not be appropriate in the context of the paper and removed it but the reader may still consider the following. I wrote: “Everyone would gain from more rational behaviour, including monopolies. Just consider pharmaceutical industries which spend twice as much on marketing than on research and development (Gagnon & Lexchin, 2008). The pharmaceutical industry saves millions of lives and is of particularly important service to society. Monopolies certainly don’t behave in a wasteful and inefficient manner by malicious intent but act within the framework shaped by policy makers and not least the work of economists. “Losses associated with the pursuit of profit provoke ethical considerations in public debate and also among economists. I guess my point was that only when you understand that there is an underlying economic problem you are able to search for and accept solutions to the problem. It is important to not attribute the cause of some problem to some perceived maliciousness which is just our way of determining that there actually is some problem. Of course, ignoring the problem isn’t helpful either. In the current framework certain common market scenarios allow producers to effect inefficient prices. These inefficiencies cause losses rightly identified as something “bad”. The pharma industry was a nice example because due to patent monopolies those losses are relatively large and at the same time people have never been healthier thanks to the advances of the pharmaceutical and chemical industries, which most people agree is a very “good” thing. In the manner we have moved from evil daemons causing our illnesses to an understanding of the human body and modern pharma-industry there is hope that we will understand our society better. E.g. the main culprit for the 07/08 subprime crisis in public debate and also economists is human “greed“. What happened is that banks lent vast amounts of money to borrowers who were in no position to repay their loans. I think this behaviour may be explained when you consider that lending to risky borrowers is simply disproportionately profitable for banks; of course, only until they actually default. This seems like a variation of the market power problem and it would certainly be interesting to scrutinize the effects of the tax-condition in that context.
Point 3: I agree. I hope, also with regard to point 4, the excel-tables give an effective example for readers.
Point 4: The linear demand and quadratic cost functions used in the set-up are the ones used in the figures of the paper. I prefer the shorter paper without closed form examples.
Point 5: Consider it done.
Point 6,7: The tax is not meant to turn into a subsidy. If you are referring in point 7 to further discussion of what happens near Q_c I have spent some time in the paper on it but it really is only a special case, e.g. an actual zero-rate seems rather unlikely in reality. If you refer to the whole analysis please be more specific; I hope I have explained everything in a comprehensible fashion.
Point 8: I think it was unnecessary to focus directly on economy-wide application. E.g. it ignores the fact that for companies of certain revenue tax-audits and accounting efforts are more elaborate so it’s obvious that tax-evasion and information questions are much different ones compared to, e.g. small businesses in monopolistic competition. Economy-wide application was important to me not only because we generally have monopolistic competition throughout the economy but also things like tax-equality; or if you assume the condition actually works and induces efficiency it would give firms operating under the condition a sort of competitive advantage compared to those operating unmitigated potentially creating unwanted distortions.
In general with regard to the applicability part I was presented with the difficulty of actually having to spell out important implications and considerations to be made even if they seem “ambitious” and not exaggerate or, worse, being led to neat but ultimately meaningless speculation. I hope I have managed to stay objective and sober.

Tim Worrall - Comment
November 17, 2008 - 10:29

see attached file

Goetz Seißer - Response
November 24, 2008 - 01:44

Thanks for your comment. Needless to say that I was previously unaware of Glaister’s work and your comparison of my scheme with Ramsey pricing is also much appreciated. Indeed, “The challenge is to know if it is possible to design an output related profits tax that can overcome some of ...[more]

... these difficulties and be better or more cost effective than other regulatory schemes.“ You say, implementation of my scheme “requires knowledge of demand and cost functions”. However, isn’t the information collected for accounting purposes, in principal, sufficient for my scheme? Naturally current accounting standards are not designed with the purpose of making a conditional tax work. It appears that nothing which is generally impractical to obtain, like demand elasticities, has to be available to authorities in order to achieve results. Together with the result of an unambigious improvement in welfare in normal circumstances this would make a stronger case for practical application. E.g., assume an economy where no corporate taxes are collected. Wouldn’t you expect the introduction of a gross-profit conditional tax - say with contemporary technology, rule of law etc. of a developed country - to be of present value to the economy?

Anonymous - Referee Report
November 17, 2008 - 13:30

see attached file

Goetz Seißer - Response
November 24, 2008 - 01:45

Thank you very much for refereeing my paper. I have revised the statement you refer to. As for a theoretical competitive optimum for a theoretical monopoly being remarkable, I’ll leave that for the reader to decide.

Anonymous - Discussion Paper 2008-28
December 13, 2008 - 23:03

See attached file.

Anonymous - Referee Report
January 02, 2009 - 12:02

see attached file

Goetz Seißer - Response
January 04, 2009 - 13:34

Thank you for your report. However, it left me puzzled on how you got the idea that "the proposed tax seems only to apply to the case of a natural monopoly". I concerned myself with natural monopolies exactly once in paper and the context was to highlight a result implicit ...[more]

... in eq. 11, since natural monopolies are not profitable at the competitive output the tax-condition cannot possibly lead to a competitive output. The tax will lead to an output decision somewhere beyond the monopolistic quantity. The exact optimum depends on the specific scenario, and you can also think about the modifier-factor to bring the conditioned profit maximum closer to the break even point. Contrary to your claim that it is focused on natural monopolies the tax-condition is, in my opinion and this is also stated in the paper, applicable to almost all market scenarios. In competition it doesn't take any effect and whenever there is some form of inefficient price setting it would dynamically counteract such behaviour, just as i thought i had explained in the paper. That is actually a very important and appealing feature. Take, for example, a cartel. Rather hard to prove and every traditional regulatory measure always represents a severe interference which eventually may be as harmful as the cartel itself, the extreme case being a planned economy. The tax-condition, however, doesn't seem to interfere with the profit incentive and would still discourage inefficient pricing. No need to proof collusion or any of the complications when dealing with cartels since the argument is clearly: If you're not doing anything wrong, it doesn't hurt you and if you are, well, with the rule you're certainly less likely to do so in the future. In case of a cartel it would already be enough if the tax discourages the illegal collusion and let real competition take over in order get rid of the losses. Clearly, your job is to stop me if I'm getting something wrong here. So if you think that the tax-condition is only applicable for natural monopolies please provide some arguments and otherwise I'm a bit worried about the comprehensibility of my paper if you missed such important aspects.