jump to navigation

Market failures, part VII maj 9, 2008

Posted by Fredrik Gustafsson in Nationalekonomi, _In English_.
1 comment so far

Diskussionen kring skalfördelar fortsätter. Nästa inlägg kommer att handla om externa skalfördelar.

Economies of scale, continued

Yet another market failure that is thought to arise partly due to the presence of increasing returns to scale is the one that is commonly thought to justify strategic trade policy. The two most famous strategic trade policy models have been developed by James Brander and Barbara Spencer[1] and by Paul Krugman[2]. The critical assumption in both models is that international trade is characterized by imperfect competition. More specifically, that an industry is dominated by two oligopolistic firms both behaving like Cournot duopolists.

The difference between the two models is that Brander and Spencer assume that both firms earn all their profits from exporting to a single third country, whereas Krugman introduces market segmentation in his model. In the Brander Spencer model, the state can alter the balance of power in the game by granting the home firm an export subsidy, making it a Stackelberg leader. This is done by making the domestic firm’s threat of increasing its output (so it will correspond to the Stackelberg leader’s position) credible. This will increase the market share and profits of the domestic firm.

In Krugman’s model, the state can shift the reaction curve of the local firm outward by giving the domestic firm a privileged position on the home market through, for example, and import tariff on the good in question. Due to the presence of economies of scale, this will set a circular causation in motion. The domestic firm will increase its market share in both the protected and unprotected market due to the lower and lower marginal costs that follow from the increase in production and the increase in market share (first directed at the home and then at the foreign market)[3].

These “market failures” are however of no interest to this paper. In fact, according to the definition used above they are not even market failures. There are no inefficiencies, neither static nor dynamic, in these models. In fact, successfully fixing this “market failure” will reduce static efficiency and at best leave dynamic efficiency unaffected. Instead, what these models suggest is that strategic trade policy can be used to shift profits from some firms (preferably foreign) to others (preferably domestic). I.e. it is in essence a way for the government to reward its friends and punish its enemies.

Regardless of whether or not this is something we think that the government should be doing, there are a number of arguments that can be made against these models. For example, it is doubtful whether there are any industries at all in the real world that fit the model. Most industries are made up of more than two firms. And even if it is possible to find such an industry, none of the two firms will be wholly or even primarily owned by individuals or organizations from one particular country. It makes no sense for, say, the US government to transfer profits from a firm owned by Americans, Japanese, Germans to another firm owned by Americans, Japanese, Germans. The models also, among other things, assume that rival governments are passive, which is totally unrealistic. The bottom line is however that this kind of strategic trade policy can in no way be used to increase dynamic efficiency, and therefore, it is of no interest to this paper.

[1] James Brander and Barbara Spencer: ”Tariffs and the Extraction of Foreign Monopoly Rents and Potential Entry,” Canadian Journal of Economics Vol. 14 (1981), pp. 371-89; ”Tariff Protection and Imperfect Competition,” in Henryk Kierzkowski (ed.), Monopolistic Competition and International Trade (Oxford: Clarendon Press, 1984), pp. 194-206; ”Trade Warfare: Tariffs and Cartels,” Journal of International Economics Vol. 16, No. 3-4 (1984), pp. 227-42; ”Export Subsidies and International Market Share Rivalry,” Journal of International Economics Vol. 18, No. 1-2 (1985), pp. 83-100; and ”International R&D Rivalry and Industrial Strategy,” Review of Economic Studies Vol. 50 (1983), pp. 707-22.7

[2] Paul Krugman, ”Import Protection as Export Promotion: International Competition in the Presence of Oligopoly and Economies of Scale,” in Kierzkowski (ed.), Monopolistic Competition and International Trade, pp. 180-93.

[3] For an excellent summary of both models, see Klaus Stegemann, “Policy Rivalry among Industrial States: What Can We Learn from Models of Strategic Trade Policy”, International Organization, Vol. 43, No. 1 (1989), pp. 73-100 and also Jan Andera, Driving under the Influence: Strategic Trade Policy and Market Integration in the European Car Industry, Lund Studies in Economic History, No. 42 (Lund: Almqvist and Wiksell International, 2007). For an introduction to the fascinating world of game theory and Cournot duopoly games Stackelberg leaders and followers, and so on, see Hal Varian, Intermediate Microeconomics: A Modern Approach, Sixth Edition, (Berkeley, CA: University of California, Berkeley, 2002).

Market failures, part VI maj 7, 2008

Posted by Fredrik Gustafsson in Nationalekonomi, _In English_.

I denna del fortsätter diskussionen om skalfördelar. I nästa del kommer strategisk handelspolitik att tas upp.

Economies of scale, continued

There are other, less extreme examples of economies of scale as well; cases where there are economies of scale up to a smaller level of production and, therefore, where the industry can still be made up of a fairly large number of firms that all produce at an efficient scale. From a dynamic efficiency point of view, the most interesting economies of scale are those that apply to discovering or developing new technologies, products are resources.

One activity that comes to mind, and that is coincidentally regarded as being associated with significant economies of scale, is research and development (R&D). Some for instance claim that, since R&D has to be carried out on a very large scale in order to be profitable, small firms cannot “afford” these investments in R&D. Hence, new technologies, products and resources are not developed and the dynamic efficiency of the economy is hampered[1].

This is however a very poor excuse of an argument. Economies of scale will never deter private firms. Economies of scale just mean that the investment has to be of a certain size in order to be profitable. As long as the investment is profitable, firms will invest. In fact, bigger is better as a larger investment means larger profits. If in fact money could be made by investing in R&D, and if the only thing preventing established firms from doing so is their own fear of large projects, someone else will make the investment, put the established firms out of business and make them regret their cowardice. Because, assuming that there is a working credit market (which we do assume) a small firm will always be able to finance large profitable investments through debt.

Furthermore, if it is profitable to carry out R&D if and only if one single firm or institute carries out all R&D, what prevents the firms in the relevant industry from forming a joint company with the specific purpose of carrying out R&D that will be made available to every individual owner of the company? The claim that economies of scale will deter private firms from investing in, for example R&D, is simply ludicrous.

Although this might be straying from the topic a little, a very similar argument is often made for risk; that the presence of large risks means that a market failure that might reduce dynamic efficiency could be created. For example, due to the fact that R&D is such a risky undertaking—where little of value can be salvaged if the research fails to yield any results—a small firm might not be able to “afford” taking the risk of investing in R&D[2].

The argument is then that by funding or undertaking R&D, the state can spread the risk more evenly across different firms and consumers in the economy. Through this, it might be able to achieve a more rapid rate of development. A similar argument is that private firms are “too” risk averse, which slows down the rate of development as “not enough” resources are devoted to discovering new technologies, products and resources. Therefore, some claim, that the government should in some way encourage firms to take greater risks or that the government should bear the risk for them.

It is however not difficult to see that these arguments are flawed as well. If the expected returns of investing in R&D are large enough, but the firm hesitates because there is a small chance that the investment will not make any profits, there is a great number of methods available on the free market that the firm can use to divest itself of some of the risk.

Furthermore, there is also a great risk involved in passing up an investment opportunity; in doing nothing. A firm cannot expect to continue to earn money by producing the same thing it has always produced, in the same way it has always produced it. A risk-averse firm might just as well invest in R&D or in a new form of production to minimize the risk that it is left behind when rival firms make important discoveries and upsets the status quo. The best defense is after all a good offense.

Lastly, it is unclear whether risk-averseness in itself can be considered as a source of market failure or that individuals can be “too” risk averse. The reduction of risk is an economic good like any other. Confronted with two projects with equal expected rates of return, but where one project is associated with significantly greater risk, the firm might be willing to pay to gain access to the second project.

For example, it is probably not very far-fetched to assume that a person would be willing to sell a lottery ticket with a 50 per cent chance of winning ten million dollars for a sum considerably less than five million dollars (which is the expected return of the ticket). To call this action a market failure just because the consumer attaches a value to the reduction of risk is clearly erroneous. As Demsetz writes:

Once it is admitted that risk reduction is […] an economic good, the relevant question for society is what real institutional arrangements will be best suited to produce risk reduction or risk shifting […] [T]he taste for risk reduction must be incorporated in the concept of efficiency […] Given the fact of scarcity, risk reduction is not available at zero cost, so that the risk averse efficient economy […] does not produce a complete shifting of risk but, instead, it reduces or shifts risk only when the economic gain exceeds the cost[3].

That is, firms may avoid investing in risky projects because they value reducing risk in the same way as consumers will buy red cars because they prefer driving red cars. For the government to intervene case because entrepreneurs are “too” risk-averse is as unjustified as the government intervening because consumers prefer red cars, when in reality green cars are much nicer. If one uses a subjective theory of value, entrepreneurs, investors, and consumers cannot be “too” risk averse any more than blue cars can be said to be objectively more beautiful than red.


[1] Grabowski, “The Successful Developmental State”.
[2] Kenneth Arrow, “Economic Welfare and the Allocation of Resources to Inventions”, Nelson (ed.), The Rate and Direction of Inventive Activity, (Princeton, NJ: Princeton University Press, 1962).
[3] Demsetz, “Information and Efficiency”.

Market failures, part V april 28, 2008

Posted by Fredrik Gustafsson in Nationalekonomi, _In English_.

I denna del kommer vi till slut in på saker som kan få marknaden att misslyckas. Först ut är skalfördelar som kommer att behandlas i en rad inlägg framöver. I och med att det är Valborg på onsdag kommer nästa inlägg först på fredag.

What can make the market fail?

In general, there are at least three things that can cause market failures: economies of scale; imperfect information; and externalities[1]. The following sections will in turn go through each of these three reasons for why the market might fail. Market failures that arise due to imperfect information and economies of scale will however only be briefly summarized.

The argument below will be that, if there are indeed any market failures that might reduce dynamic efficiency and that the government has a reasonable chance or resolving, they primarily arise due to externalities. After summarizing market failures arising due to imperfect information and economies of scale, the remainder of this article will therefore solely focus on market failures that arise due to externalities.

Economies of scale

Economies of scale have always been one of government’s favorite reasons for intervening in the economy and their popularity as an excuse for government intervention has persisted to this day. But does the claim that the market can fail due to economies do scale have any theoretical merit? Can economies of scale reduce an economy’s dynamic efficiency and, hence, its long-term economic growth? And if this is truly the case, can we expect that the government will be able to do anything about it?

Before anything else, economies of scale need to be defined since there are in fact two different kinds. First, there are economies of scale in the normal sense of the word: that is, a firm’s average cost of production decreases as the production of the firm increases (up to a certain level of production). This situation usually arises due a presence of large fixed costs. This is what we call internal economies of scale.

The other kind of economies of scale are, hardly surprising, called external economies of scale. They mean that the costs of the firm decrease (or that the revenues increase) as the total output of the industry (or other industries) increases. These economies of scale are also referred to as complementarities, economies of agglomeration, network effects, or whatever term might be in vogue at the moment.

But let us start with internal economies of scale. The most extreme case of economies of scale is when the increasing returns to scale continue beyond the production level that is equal to the total supply of the good on the entire market. This means that, since the economies of scale are so large, the lowest average cost per produced unit can only be attained if one single firm produces the entire supply of this particular good. In other words, the production can only be X-efficient if the industry is characterized by a true monopoly where one single firm produces all the output.

This very unique situation is called a natural monopoly. Libertarians usually take offense to the very term natural monopoly since they are of the opinion that monopolies only can exist by the virtue of government regulation or protection. However, the fact that a certain industry could be described as a natural monopoly does not necessarily mean that one single firm will dominate the entire industry. It only states that, in order for the industry to be X-efficient, the entire output has to be produced by one single firm.

Natural monopolies are usually thought to create market failures since it does not matter, it is claimed, whether one or several firms produce the product: the production will always be statically inefficient. Why this is the case if two firms or more produces the product is easy enough to understand since it means that the product is not produced as cheaply as it could be produced if the entire output was produced under one roof.

Whether a market monopoly automatically leads to inefficiencies is a trickier question. If we assume that the monopolist is actually able to restrict its production, and through this, increase prices, without a competitor entering the market, neoclassical economics teaches us that there will be inefficiency. This is because the monopolist is able to transfer some of the consumer surplus to itself while some of this surplus is lost due to the fact that production is restricted.

Now, it is indeed very temping to say that this situation is inefficient, especially if one looks at the graphic representation of this situation: it is perfectly clear that the area on the graph that represents the gain in surplus of the monopolist is smaller than the area in the same graph that represents the surplus loss of the consumers.

The problem is that we cannot compare producer and consumer surplus in this way. We know that the consumers experience a loss in welfare and the monopolist experience a gain in the same. We can also say that there is a deadweight loss (in welfare) due to the fact that the monopolist is successful in restricting production. We can even express these gains and losses in Dollars, Euros, blueberries or whatever.

The one thing we cannot do is to say whether the psychic loss of the consumers is in fact larger than the psychic gain of the monopolist. Value is purely subjective, and it is impossible to tell whether the psychic loss of the consumers is actually greater than the psychic gain of the monopolist. For all we know, the monopolist might attach extreme value to his surplus gain, whereas the opposite is true for the consumers.

At most, one can say that it is possible (partly due to the deadweight loss) that this monopoly leads to a real inefficiency. Perhaps one could even go as far as saying that it is likely that it leads to inefficiencies. But there is no way to say for certain that natural monopolies lead to static inefficiency. And besides, even if they do, it is clear beyond doubt that they do not reduce dynamic efficiency, and for this reason natural monopolies are of no interest to this article.

[1] Grand, “The Theory of Government Failure”; Gómez-Barroso and Pérez-Martínez, “Public Intervention in the Access to Advanced Telecommunication Services”.

Market failures, part IV april 25, 2008

Posted by Fredrik Gustafsson in Nationalekonomi, _In English_.
add a comment

Nu vet jag inte om det faktum att det inte blivit någon debatt runt senaste inlägget beror på att det jag skriver är så förbaskat övertygande eller att ingen läser det. Hur som helst kommer här nästa del som gör det lite tydligare vad vi egentligen menar när vi säger att ”marknaden” misslyckas.

Can markets really fail?

Before we turn to the important question of what can make the market fail, we have to ask ourselves if we really can say that the market fails. That is, if speaking of market failures makes any sense. In order for there to be failure, there has to be action. If someone or something does not act, it cannot fail. But the market cannot act; only individuals can act. This would in turn mean that the market cannot fail and that it makes no sense to talk about market failures.

The argument is that when we say that the market fails to produce at the lowest cost, to allocate resources in the most efficient manner and to develop new technologies, industries, products and resources, what we really mean to say is that individuals on the free market fail at doing these things. Is it not the owner of the factory that uses too much resources when producing his good, the investor that invests his money in ventures with low returns and the inventor that does not succeed in developing that new technology, even though it is within his capability (or at least physically possible) that fail, and not the free market? And no one would argue against the notion that individuals can fail.

At first glance, this objection seems reasonable. On closer inspection, it turns out that it is pure semantics and of no or little consequence for our discussion. What is meant by the (free) market when market failures are discussed in this article is a set of formal and informal rules that govern and restrain human action. The term does in this case not refer to a place where goods or services and bought or sold. That is, when we say that the market fails, we mean that a certain set of rules fail to produce either maximum static or dynamic efficiency—or, fail at producing higher static or dynamic efficiency than a different set of rules.

To call this set of rules that govern the behavior on a free market for “the free market” is perhaps somewhat sloppy. The more proper way to refer to the thing we mean by market failures is “a situation in our laissez-faire economy where static or dynamic efficiency is lower than an imagined optimum, or, the outcome that would follow from government intervention”. But simply calling these situations market failures is of course a lot easier, even though the term can be somewhat confusing.

Take the way we speak of the “failure of socialism” as an example that proves that this objection is rather silly. Few would argue that we cannot say that socialism has failed to produce static and dynamic efficiency. No sane person would argue that it was not socialism that failed, but instead individuals in socialist countries that failed to produce their products at a low cost, allocate resources to where the returns were highest and develop new technologies, products, industries and resources. In this case, it should be clear that it was the set of rules that governs and restrains human action in socialist countries that failed to produce efficiency, not the individuals that actually produced the expensive products that no one wanted.

Sure, ultimately it was the individuals in the Soviet Union that acted inefficiently (as socialism cannot act). But such a claim misses the point that, while part of the reason for why they acted inefficiently might have been their own incompetence, the main reason for why they acted inefficiently was due to the fact that the rules imposed upon them by the state either prevented them from, or failed to produce sufficient incentives for, acting in an efficient manner. In the same way, this article will study whether the set of rules that govern and restrain behavior in the “perfect” laissez-faire economy too can prevent and fail to provide sufficient incentives for economic growth.

Market failures, part III april 23, 2008

Posted by Fredrik Gustafsson in Nationalekonomi, _In English_.
add a comment

För det första vill jag påpeka att jag lagt till ett flertal stycke på slutet av det förra inlägget där jag lite mer noggrant förklarar hur statisk och dynamisk effektivitet interagerar. Jag kan verkligen rekommendera att ni tar en titt: de grejer som jag lade till ökade verkligen min förståelse av dynamisk effektivitet. I denna tredje del börjar det hetta till på allvar när marknadsmisslyckanden skall definieras. I och med att denna del är rätt lång, och att det säkert finns en del att diskutera, kommer jag först att publicera nästa del nästa vecka eller när all eventuell diskussion ebbat ut. I nästa del kommer jag att diskutera om vi verkligen kan säga att det är marknaden som misslyckas (och inte individer på marknaden), så Kaj kan spara det argumentet tills dess.

Defining market failures

The question then becomes: how does one define a market failure? The question seems straightforward enough, but unfortunately, the same cannot be said for the answer. In fact, much of the confusion that surrounds the idea of market failures is in part created by the fact that the concept can be defined in at least two different ways and that people have a tendency to confuse the two definitions. Before the term market failure is defined, something of great importance to the discussion that will follow however has to be pointed out.

The starting point will in this article be the ideal night-watchman state that already has all the institutions thought to be necessary for creating economic growth. That is, the starting point is an economy where the government’s only role in the economy is to enforce property rights and contracts—a role it plays to perfection—and an economy that has fully developed markets and a stable currency.

For an economist, at least with a classical liberal disposition, it is the perfect economy and there is, prima facie, nothing standing in the way of economic growth. The question this paper will ask is thus whether there is anything that can cause market failures in this perfect economy; and, whether such an economy can ever increase its level of growth by government intervention in the form of tariffs, subsidies, industry regulations, and so on.

The first and most commonly used definition of a market failure is a situation where, in this perfect laissez-faire economy, a change that would potentially leave everyone better off is, for some reason, not undertaken. A very simple example of this is when a voluntary exchange between two individuals, which would leave them both better off, is for some reason not undertaken. As the exchange is not made, they are both worse off compared to if the exchange would be been made.

In general, the market can according to this definition fail in three different ways: it can fail to produce maximum X-, allocative and dynamic efficiency. That is, if it is physically possible to produce a given product that is currently being produced on the free market using less, labor, time or natural resources we are dealing with a market failure. Similarly, if it is physically possible to reallocate the use of resources in a way that is more effective in fulfilling human wants, we are dealing with a market failure.

And lastly, if it is physically possible to develop a new technology, industry, product or resource that would potentially leave everyone better off, we are dealing with a market failure. Thus, whenever the static or dynamic efficiency is lower on the free market than the highest possible static or dynamic efficiency imaginable (so long as it is physically possible given available resources and technology), the free market is said to be failing.

The first thing that can be noted concerning this definition is that it is very generous. It implies that we are constantly surrounded by an almost infinite number of market failures. At every single point in time, every product that is produced on the free market could probably be produced a little bit cheaper; every resource in use could probably be used in a slightly different manner in order to render a greater utility to its user; and at every point in time, there are surely countless new (profitable) technologies, industries, products and resources that could be developed, the only problem being that no one has yet to figure out how.

So, defining market failures in this way means that the market fails in relation to an ideal outcome where every individual is completely rational, has perfect information, and so on and where no one could have acted any differently without at the same time decreasing his utility. To use a neoclassical term, according to this definition the market fails in relation to the situation where perfect competition prevails.

This is the definition of market failures that is most commonly used, and although it might seem to be way too abstract to be of any practical use, it is in fact of great theoretical interest. By definition, the government can only increase efficiency by intervening when the efficiency we can observe on the free market is not equal or close to this “maximum” efficiency. This follows logically. As will hopefully be shown below, our best chance of finding situations where the government can in fact increase dynamic efficiency by intervening is by first identifying situations where the dynamic efficiency that prevails on the free market is, for some reason, significantly lower than this maximum efficiency.

Although this definition can certainly be useful, it has unfortunately been terribly misused by some. The most fatal error one could make as regard to this definition is to simply assume that the fact that we can identify situations where the outcome on the free market is sub-optimal compared to this “ideal” outcome automatically means that the government can increase efficiency by intervening; that the presence of market failures according to this first definition implies that the free market is inefficient compared to government intervention. It cannot be pointed out enough how gross of an error this is.

According to the above definition, the market fails in relation to an imagined optimal outcome. In no way is it implied that we can actually know what this optimal outcome looks like or that it can be reached through government intervention. If the people that worked for the government were omniscient and only had incentives to do what is best from an efficiency point of view, this would perhaps be the case. But no one, not even the people that are most enthusiastic about government intervention, would think of claiming such a thing. It is therefore of the utmost importance to point out that the existence of these kinds of market failures alone can never warrant government intervention.

This discussion brings us to the second definition of market failures. According to this definition, the market does not fail in relation to an imagined outcome. Instead, the market is said to fail when the outcome on the free market (in terms of static or dynamic efficiency) is inferior to outcome that would follow from government intervention. That is, the market fails when government intervention would mean that products would be produced using less resources, that resources would be allocated to production fulfilling more urgent wants, or that government intervention would lead to new technologies, industries products and resources, that would potentially make everyone better off, being developed.

And it is important to remember that the “outcome that would follow from government intervention” is not some hypothetical outcome that an omniscient and benevolent government could produce, but the actual outcome a certain government, with all its flaws, will in fact produce. This definition of a market failure is therefore dependent on, not only a theory of how the market works, but a complete and correct analysis of government action and failure, not on fantasies and wishful thinking concerning what the government could do if only it acted the way you wanted.

It should now be clear why the fact that we can define market failures in these two very different ways has created a lot of confusion. For example, when someone that favors government intervention says that we are at all times surrounded by countless market failures, he is most likely referring to the first definition. And there is no arguing here: if this definition is used, the number of market failures on a free market is indeed infinite. This does not however in any way mean, by itself, that efficiency could be increased by intervention or that the free market is inferior at creating efficiency than any other economic system imaginable.

Unfortunately, those who favor government intervention often fail to realize this. They often incorrectly assume that the fact that we can observe inefficiencies on the free market in comparison to a “perfect” outcome means that we actually can fix them through government intervention. In essence, they are guilty of the same mistake Sidney Hook talks about in his famous quote: “I was guilty of judging capitalism by its operations and socialisms by its hopes and dreams”. They compare the actual outcome on the free market, not to the actual outcome that would follow from government intervention, but to an ideal outcome that only exists in our imagination.

Similarly, when libertarians confidently assert that there are no such things as market failures they are referring to market failures according to the second definition. That is, they are not saying that the free market always produces the best imaginable outcome every single time, but that the actual outcome that follows from laissez-faire is always superior to that which follows from government intervention. Thus, it is perfectly possible that there are countless market failures according to the first definition at the same time as there are no market failures according to the second definition.

Market failures, part II april 21, 2008

Posted by Fredrik Gustafsson in Nationalekonomi, _In English_.

Den andra delen i min serie om marknadsmisslyckande kommer, likt den första, inte att innehålla så mycket kontroversiella saker. Istället fokuserar jag på att definera effektivitet, vilket är nödvändigt för den fortsatta diskussionen. I nästa inlägg, som publiceras på onsdag, kommer jag att komma in på definitionen av marknadsmisslyckanden. Jag kommer att försöka publicera alla inlägg i framtiden på måndag, onsdag, fredag så att ni är mentalt förberedda. 🙂

Static and dynamic efficiency

The free market is said to fail when it creates inefficiencies. But what is efficiency? In general, one can say that there are two basic types of economic efficiencies: static and dynamic efficiency. There are furthermore two kinds of static efficiencies: X- and allocative efficiency[1]. X-efficiency is what a layman would normally regard as efficiency: that is, when something is produced at the lowest possible cost. Or, to put it differently: when a product is produced using the least amount of resources possible.

If we assume that someone engages in production, it is a universal fact that this person will try to produce the maximum output of goods per unit of time from given units of factors[2]. It is always true that the person will prefer producing the same amount of good using less time and/or resources, since he can use the time and/or resources left over in order to satisfy less urgent wants. In other words, given that man engages in production, he will always prefer higher X-efficiency. If confronted with two possible ways to produce a certain product, where the only difference is that the second way means that the exact same product can be produced using less time, labor, and so on, he will always prefer the second way.

If X-efficiency is what laymen mean by efficiency, allocative efficiency is what economists mean by efficiency. In a sense, while X-efficiency means that resources are used efficiently, allocative efficiency means that they are used effectively. That is, allocative efficiency means that the resources are dedicated to the production that satisfies the most urgent and unfulfilled human wants. If the resources could be used to fulfill a more urgent want or more wants than they are currently fulfilling, their use is not allocatively efficient. This is the same thing as saying that the resources are not allocated to the activity with the highest private rate of return: that the owner of the resources could earn higher returns if he reallocated his resources.

The difference between these two efficiencies can be illustrated with a very simple example. Imagine Robinson Crusoe stuck on a deserted island. In order to survive, Robison spends his time picking blueberries. Say that he manages to pick ten kilos a day. Now, imagine that Robinson could use a different method for picking blueberries that would allow him to pick twice as many blueberries while not expending any more time and effort. If Robinson, for some reason, still uses the labor- and time-consuming method, his blueberry picking is X-inefficient. And Robinson would prefer using the X-efficient method. Using this method, he could either double his daily harvest of blueberries (without exerting any more effort), or, spend only half the normal time and effort picking blueberries, enabling him to use the extra time and energy to perhaps pick mushrooms or rest. Since Robinson obviously prefers more blueberries over less and more leisure over less, he obviously prefers the more X-efficient way of picking blueberries.

To see how allocative efficiency works, imagine that Friday would be willing to exchange ten kilos of mushrooms for 40 kilos of blueberries. Say also that Robinson could pick ten kilos of mushrooms using the exact same time and effort he uses when picking ten kilos of blueberries. In this situation, Robinson could in fact increase his production of blueberries fourfold by simply picking mushrooms and exchanging them for blueberries, without increasing the effort or time spent working in the forest. If for some reason Robinson continues to pick blueberries, his use of resources is clearly allocatively inefficient. If he was to reallocate his time and effort to picking mushrooms, he could either significantly increase the amount of blueberries he could consume or increase his leisure.

Even tough this paper will touch upon static efficiency time and time again, the main goal is to identify market failures that may reduce the dynamic efficiency. But how does one define dynamic efficiency? As will be shown shortly, an economy’s dynamic efficiency is equal to its ability to generate lasting economic growth[3]. The more dynamically efficient an economy is, the higher its economic growth.

This can perhaps best be shown through another Robinson Crusoe example. Say that Robinson’s blueberry picking is already statically efficient. How can he in this situation further increase his income or well-being? First, he can increase the resources devoted to production. In the above case, he can work more. Secondly, he can try to devise an entirely new method of picking blueberries that will enable him to pick more berries while not exerting any more time or effort. Perhaps try to invent some sort of blueberry picking machine. Thirdly, he can try to discover something he (or Friday) values even higher than blueberries (at least relative to the effort that goes into producing it). Perhaps there is some other berry in the forest that is even tastier or easier to find than blueberries, but that no one has thought of picking yet. Or perhaps Robinson can try to invent an entirely new product. Say he starts making sculptures instead and finds out that Friday is willing to pay 100 kilos of blueberries for a single sculpture, even tough it takes Robinson no more time or effort to make this sculpture than to pick ten kilos of blueberries.

What this simple example shows is that the only way to produce long-term increases in income (economic growth) is to develop new technologies that makes production cheaper, new products that better satisfy the wants of the consumers, or to discover new, valuable resources that can be used in production. I.e. a dynamically efficient economy is an economy where such new technologies, products and resources are continually discovered or developed, which will allow for a steady increase in well-being.

A couple of things however have to be pointed out. First of all, it should be remembered that a high dynamic efficiency, at least as far as the development of new technologies and products are concerned, manifests itself though continuous increases in static efficiency. To see this, imagine first an economy in a general equilibrium that is perfectly X- and allocatively efficient. While this economy is statically efficient, there is no growth; the economy is perfectly static and in a stable equilibrium. Given available technology, products and resources, there are no further gains in efficiency or income to be made.

However, suppose that someone discovers a new technology that allows for a cheaper production of an existing product and that someone else discovers an entirely new product which is cheaper to produced or valued higher by the consumers. The very moment these discoveries are made, the economy is no longer statically efficient or in a stable equilibrium. And this will continue to be the case as long as the producers still use the old technology, which means that they are producing at a higher cost than necessary. However, as soon as the producers implement the new technology in their production, the economy once again becomes X-efficient.

The same thing is true for the case when someone discovers the new product. As soon as this new product is discovered, production is no longer allocatively efficient: reallocating resources to produce this new product, instead of an old product that the consumers no longer want given this innovation, would fetch higher returns. Thus, only when the producers begin to produce this new product will the economy be once again be allocatively efficient and equilibrium will be restored.

What high dynamic efficient thus entails is that the equilibrium in the statically efficient economy is continuously upset. If the dynamic efficiency is high enough—i.e. if these types of discoveries are made often enough—the economy will in fact never be in equilibrium. This economy will always move towards what is X- and allocatively efficient, but because of the high dynamic efficiency, what is X- and allocatively efficient changes so often that the producers are never really able to catch up. They are like the donkey that has a carrot dangled in front of it. It runs and runs towards the carrot, but it never quite catches it. However, like the donkey, the economy is continuously moving forward, towards increased efficiency.

Thus, while the dynamically efficient economy may never be statically efficient, it continuously moves toward a higher level of static efficiency and gets more and more efficient along the way. The new potential equilibriums created by the high dynamic efficiency are always on a higher level in terms of static efficiency. That is, even though costs are never as low as they could be and resources are never allocated to where they generate maximum returns, costs are continually lowered and resources are continually being reallocated to production that better satisfies the wants of the consumers. This is how a high dynamic efficiency creates lasting economic growth.

The most proximate cause of the increase in income is that entrepreneurs try to increase static and dynamic efficiency and move the economy towards equilibrium given these discoveries. However, without these discoveries by innovators, which create these disequilibria, the economy and the increase in income would eventually come to a standstill as entrepreneurs are successful in restoring equilibrium. One way of separating those actions that increase static efficiency from those that are a sign of high dynamic efficiency is therefore that increases in static efficiency moves the economy closer to equilibrium whereas those investments and subsequent discoveries that are signs of high dynamic efficiency moves the economy further away from equilibrium.

Another thing that has to be pointed out is the role of time preferences, which have a profound impact on how we define dynamic efficiency. Take the example of Robinson again. Say that his production is statically efficient and that he now contemplates how to further increase his income. As noted above, he can either try to develop a new technology or product, or, he can work more. Say that he under no circumstances wants to work any more than he is currently doing. His options are therefore to either try to develop a new technology that will make his blueberry picking more efficient, or a new product that he, or Friday, values higher.

The problem is that trying to develop this new technology or product requires effort. Since Robinson does not want to work any more than he is currently doing, his only choice is to work less at collecting blueberries while he tries to develop the new technology or product. This in turn means that he has to restrict his current consumption of blueberries (or the thing he gets in exchange for these blueberries) while he tries to develop the new product or technology. This is in fact the very definition of what an investment is: restricting one’s current consumption in order to increase one’s future income.

Therefore, trying to develop new technologies, products or resources always entail restricting current consumption in order to increase future income, regardless of whether we are talking about Robinson Crusoe or a modern economy. This is where time preferences come in. It is a universal fact that if man has a choice between consuming a product now and consuming the exact same product later, he will always prefers to consume the product now. Murray Rothbard describes this universal fact of time preference in the following way:

At any point of time, and for any action, the actor most prefers to have his end attained in the immediate present. Next best for him is the immediate future, and the further in the future the attainment of the end appears to be, the less preferable it is. The less waiting time, the more preferable it is for him[4].

Therefore, restricting current consumption is always a real cost and has to be weighed against the returns to the investment. Furthermore, while it is universally true that man prefers consumption in the present over consumption in the future, the intensity of the time preference is bound to vary greatly among different individuals. For those whose income is so low that they cannot restrict their current consumption at all without starving to death, the cost of restricting current consumption is enormous.

But what does this have to do with dynamic efficiency? The problem is that if we take this universal fact of time preferences into account, and also the fact that the intensity is bound to vary among individuals, we cannot say for sure whether Robinson is behaving in a dynamic inefficient manner if he does not devote any time or effort to developing new technologies or new products, even if these efforts would mean he could greatly increase his production. It might be the case that Robinson (who is not willing to put in any additional work) would not be able to survive if he spent any less time picking blueberries. It might be the case that he cannot restrict his current consumption without starving to death. And if this is the case, it does not matter how many blueberries he could pick with the new blueberry picking device since it is hard to pick and enjoy berried if you’re dead. Thus, while the returns to trying to develop a new blueberry picking device, or to find another even tastier berry, may by very high, the costs associated with doing this are simply too large.

Therefore, simply observing that Robinson, or an economy, does not invest in developing new technologies, products or resources, does not automatically mean that dynamic inefficiencies are present. An economy with no economic growth whatsoever can still be dynamically efficient. It may just be the case that it is stuck in a poverty trap where incomes are so low that no one is able to restrict his or her current consumption in order to be able to carry out investments. Therefore, the correct definition of dynamic inefficiency is when there are actors that are willing to restrict their current consumption in order to make these kinds of investments (i.e. it is privately profitable to make them), but that invest in developing a certain new technology, product or resources when there are in fact other, more profitable new technologies, products or resources that could have been chosen. Or, when there are actors that are willing to make these kinds of investments, but when the resources are, for some reason, still devoted to consumption.

According to this definition, Robinson Crusoe is behaving dynamically inefficient when he chooses to, for instance, develop a blueberry picking device instead of trying to discover a new edible berry, when the returns to the discovery of this new berry would in fact have been much larger, or, when he chooses not to invest in either of the two even though both projects are privately profitable after all costs (including those arising from time preferences) have been taken into account. Similarly, an economy is dynamically inefficient when there is a supply of capital but when this capital is devoted to developing technologies, products and resources with a very low return, compared to other technologies, products and resources that could have been developed. Or, when this capital is devoted to other activities entirely, even though the returns would have been much greater had the capital been devoted to developing new technologies, products or resources.


[1] Julian Le Grand, “The Theory of Government Failure” British Journal of Political
Science, Vol. 21, No. 4 (1991), pp. 423-442; Harvey Leibenstein, “Allocative Efficiency vs. ‘X-efficiency’”, American Economic Review, Vol. 56, No. 3 (1966), pp. 392-415; Nicholas Kaldor, “The Irrelevance of Equilibrium Economics”, The Economic Journal, Vol. 82, No. 328 (1972), pp. 1237-1255; Heinz Arndt, “’Market Failure’ and Underdevelopment”, World Development, Vol. 16, No. 2 (1988), pp. 219-229.
[2] Murray Rothbard, Man, Economy, and State.
[3] Geske Dijkstra, ”Trade Liberalization and Industrial Development in Latin America”, World Development, Vol. 28, No. 9 (2000), pp.1567-1582.
[4] Rothbard, Man, Economy, and State, p. 15.

Should the concept of market failures be taken seriously? Part I april 18, 2008

Posted by Fredrik Gustafsson in Nationalekonomi, _In English_.

Enda sedan jag skrev klart min doktorsavhandling om marknadsmisslyckanden och industripolitik i den malaysiska palmoljesektorn har jag velat skriva en artikel som tar sig an teorin om marknadsmisslyckanden ur ett klassiskt liberalt perspektiv. Till stor del för att verkligen komma fram till vad jag personligen tycker om marknadsmisslyckanden: om det är någon man behöver bry sig om eller inte. Jag håller för närvarande på att skriva en sådan artikel på engelska, och tanken är att publicera delar av den här på bloggen vartefter jag skriver den, i första hand för att få respons från andra libertarianer. I och med att den inte är klar än kan jag förstås inte säga hur många delar det blir. Men jag börjar med att posta inledningen idag, och så fortsätter jag med bland annat definitionen av marknadsmisslyckanden efter helgen.


Among liberty-minded scholars, few concepts carry the same aura of controversy as does the concept of market failure. Some libertarians even take offense to the very idea that the free market can fail. “Fail in comparison to what, the wonders of Soviet-style central planning?” they ask themselves. But who can blame them, given the way the concept of market failures has been used in the past. There in fact seems to be no end to the cases where “market failures”—the idea that the free market cannot be trusted to allocate resources or to generate economic growth—have been used as an excuse for government intervention that has always produced miserable results (except perhaps for the people in power and their cronies).

We have heard it all before. “Some tasks are simply too important to be left to the free market.” “The presence of natural monopolies necessitates regulation.” “Externalities, imperfect information, economies of scale and complementarities all mean that the market will not function efficiently, and the government therefore needs to step in”. And every single time, or at least so it seems, the promise that “everything will turn out for the better if only the government intervenes to resolve these market failures” has failed to produce anything but corruption, financial crisis and inefficient industries that crumble as soon as government support or protection is withdrawn.

Even if we disregard from the seemingly quite awful track record of government intervention designed to fix market failures, one does not have to be a genius to figure out that most arguments in favor resolving market failures through government intervention are also logically flawed to the point of embarrassment. Most of them are based on pure confusion where the “market failures” they refer to are either not market failures at all or could not possibly be fixed through government intervention.

However, if one were to peel off all the nonsense, the logical flaws, the inconsistencies and all the unrealistic assumptions that governments are benign and omniscient, and that markets are totally incapable of self-correction, would there be anything left? Is there anything worth taking seriously in the concept of market failure, or is the very concept so inherently flawed and disconnected from both logics and facts, that we can continue ignoring it. Is there anything in the concept worth salvaging? Is it possible that under some circumstances, in some places and during certain points in time, the free market will fail to generate economic growth and that the government, in these cases, has to intervene for such growth to come about?

This article will argue that it would be a mistake for libertarians, and others, not to take the concept of market failure seriously. True, few concepts have been used so recklessly as market failures. However, this is only in part due to the malicious intent of those who will grasp at any argument that favors an increase in the size of government. The most important reason is instead that market failures is a very difficult concept that few, regardless of their take on government intervention and liberty, seem to grasp.

The purpose of this article is thus to review the concept of market failure and ask a number of important questions. What is a market failure? What does the market fail in relation to? In which different ways can it fail? What can make the market fail? Are there any examples of market failures that may inhibit economic growth and where the government has a reasonable chance to resolve these market failures and thus increase growth by intervening? What would be the answer, if any, to the many different types of critiques against the concept of market failure and other concepts, such as social returns, that it rests on; to the critique that the government is inherently incapable of resolving market failures even if they do exist; that private bargaining will always resolve any and all market failures; and that government intervention that aims to resolve market failures that inhibit growth are in direct violation of private property rights?

Incentives matter juni 26, 2007

Posted by Fredrik Gustafsson in _In English_.

Dani Rodrik, one of the most promising and interesting economists at the moment, has posted a highly readable entry on aid. The book that perhaps epitomizes the second view on the plight of poor countries is William Easterly’s The Elusive Quest for Growth, I book a highly recommend. Similar reasoning can also be found in The White Man’s Burden, by the same author.

All publicity is good publicity maj 4, 2007

Posted by Fredrik Gustafsson in _In English_.
add a comment

Yesterday, I sent a press statement to the local dailies and a local radio station announcing that I will run for a seat in the Åland parliament and on the city council in my hometown Mariehamn for the Liberals on Åland. This morning, I found out that at least one of the dailies had picked up the story. The paper featured a fairly large article on my political views, classic liberalism, social liberalism and what the party president thought of my views that, to say the least, are in conflict with the party’s very social liberal profile.

That party president Viveka Eriksson of course made it clear that the party does not support some of my libertarian ideas such as a privatization of health care and schools. But she also stated that I am free to form my own opinions and that young people often have slightly more provocative view. She also made it clear that I regularly attend meetings and that I support the party’s programs. This is true, I do support the party’s views on the proposed municipality and school reforms and I am a hundred percent behind the party’s view on the language issue, which is a very important question on Åland.

In the end, I am of course glad that my announcement to enter politics was not ignored, even though the article tried to make the argument that there is a rift between me and the party, which is simply not true. I believe that both classical and social liberals can cooperate in the same political party as the two ideologies, after well, have a  lot in common . And as the saying goes, all publicity—especially if it involves a big picture of you on page six in the largest daily on the island—is good publicity.

Increasing the price of illicit drugs april 18, 2007

Posted by Fredrik Gustafsson in _In English_.

I handed in the first draft of my PhD-dissertation just before lunch today, so I am spending my “afternoon off” by reading articles on the connection between drug laws and crime. Anything to get my mind off Malaysian industrial policy.

One interesting fact in the literature is that the idea that hitting drug dealers would lead to an increase in prices of drugs has not worked out at all. While the prisons have been filled with drug dealers, prices of cocaine and heroin have been cut in half. Why? Does not an increased risk of incarceration and of loosing you’re your stash or shipment lead to higher costs?

The thing that seems to cause this curious relationship, where an increase in arrests of drug dealers actually lead to lower prices, is that these arrests will only open up vacancies in the drug trade for newcomers. The positions of arrested dealers will be filled almost immediately. And when the arrested dealers have served their time, they, for various reasons, usually decide to go back to their old profession as drug dealers and try to reclaim their old corners, even tough the market is saturated so to speak. In this way, a greater number of dealer arrests actually tend to increase competition and lower the prices.

A quote that illustrates part of this point:

If sticking a drug dealer in jail meant fewer dealers on the street, perhaps this wave of incarceration would eventually do some good. But it doesn’t work like that: Lock up a murderer, and you have one less murderer on the street. Lock up a dealer, and you create a job opening. It’s like jailing an IBM executive; the pay is good, the job is appealing, so someone will move into the office before long. Clearing dealers from one neighborhood only means they’ll move to another. Busting a drug ring only makes room for a competitor.