PAPER 1/09
Arbitrariness and expediency
Giovanni Birindelli
Measures taken by government(s), for example in order to contain and even reverse an economic crisis, may be judged according to different standards. One can be a standard based on principle: for example, “Is this measure compatible with liberty? What is my idea of liberty in the first place?”. Standards based on principles are not my concern here. A particular standard according to which the same measures can be judged is that of expediency: “is this measure likely to solve the problem at hand?”. Of course, whether a particular measure is likely to solve a particular economic problem depends, amongst other things, on the specific characteristics of that particular measure and of that particular problem, as well as on the infinite external circumstances that configure the environment in which the measure is taken. However, on a more general level, there are some abstract criteria that can be used in order to assess whether, in the long run, a measure is likely to contribute to an improvement of the situation or to its worsening. One of these, is the use of knowledge that this measure implies: in other words, the kind of knowledge it values (that is, on which it counts for the solution of the problem). In this paper I will argue, in an intuitive, informal way, that a measure (or an entire policy approach) can value either peripheral knowledge or centralized knowledge. I will call arbitrary a measure that uses centralized knowledge more than it is strictly necessary, and non-arbitrary a measure that reduces the use of centralized knowledge as much as possible. I will then argue that the quantity and the quality of knowledge valued by an arbitrary measure are very much inferior to those valued by a non-arbitrary measure and therefore that, in the long run, an arbitrary measure is much more likely to fail than a non-arbitrary one. I will leave to a further paper the argument that arbitrary measures tend to be self-reinforcing, and that they tend to destroy more and more the factors that make an economic system vital, approaching and often surpassing the point of no return. The kind of knowledge a particular measure values is not, of course, the only abstract criteria to be taken into consideration in order to judge whether that measure is likely to work or not, however I will argue that it often tends to be one of the most important ones. This is, in my opinion, one of the strongest and potentially most intuitive arguments of liberalism,and in particular of Hayek’s thought (Hayek, 1945, 1961, 1982). However, at least in my experience, this argument is rarely known or taken into consideration by those who do not study political philosophy or economics, and most of the times it is explicitly avoided by those who are in favour of collective approaches.
Peripheral vs. centralized knowledge In order to illustrate intuitively the concepts of peripheral and centralized knowledge, and those of the different orders brought about by each of them, I will start by using a metaphor, which to me it looks so simple that I would be surprised if it had not been used before. I will call this metaphor the “roundabout vs. traffic light” metaphor. Let us consider two alternative ways to regulate traffic at a crossroad: a roundabout and a traffic light. What is the main difference between them? The main difference is constituted by the kind ofknowledge they use in order to reach the same aim. The traffic light uses the knowledge of the local authority (that is, of the particular person who decided the timing of the traffic light). The roundabout, on the other hand, uses the knowledge of the driver; more precisely, of every driver who happens to cross that roundabout. The knowledge used by the person who decided the timing of the traffic light is based on his individual judgement, which depends on his own expertise, studies, preparation, experience etc. Also the knowledge used by every driver when crossing a roundabout is based on his individual judgement; however, unlike the previous case, this judgement depends on particular circumstances of time and place: if, in that precise place and on that exact moment, there are no cars approaching, the driver knows that he can safely cross the roundabout and takes the decision to cross under his own responsibility. Let us call centralized knowledge the kind of knowledge used by the authority in deciding the appropriate timing of the traffic light (that is, the kind of knowledge based on an individual judgement which does not take into consideration particular circumstances of time and place and which does not involve individual responsibility regarding the particular acts the decision refers to – if the traffic light is broken and shows green on both sides, the driver would not be responsible for the accidents that would happen as a consequence). Vice-versa, let us call peripheral knowledge the kind of knowledge used by every driver in deciding whether to cross the roundabout (that is, the kind of knowledge based on an individual judgement which does take into consideration particular circumstances of time and place and which does involve individual responsibility regarding the particular act the decision refers to). While the use of peripheral knowledge allows to grasp opportunity, the use of centralized knowledge allows only to execute commands: in fact, if the traffic light is green, by crossing I execute a command, I do not grasp an opportunity; vice-versa, if the traffic light is red and there is no other car approaching, I am forced not to grasp the opportunity of crossing (I am unnecessarily coerced). Often we underestimate the importance of opportunity. In fact, in many cases we even try to protect ourselves from it, as if it was an enemy: in other words, we frequently try to program everything, to have everything under our control, to reduce as much as possible our exposure to accident. In other words, we often tend to apply the rational methodology that is successful inside an organization also outside of it, without understanding that, in so doing, we tend to reduce, and not to increase, our chances of success or of development, and even of survival. As Hayek says, “Humiliating to man pride as it may be, we must recognize that the advance and even the preservation of civilization are dependent upon a maximum of opportunity for accidents to happen” (Hayek, 1961, p. 29). The possibility to grasp opportunity necessarily depends on the freedom to make decisions based on particular circumstances of time and place, under the decision-maker's own responsibility. The lack of this possibility implies, not only the impossibility to grasp opportunity (that is, to use precious knowledge available only to the individual), but also unnecessary coercion of the individual by the authority; in other words, it implies the absence of freedom. Leaving aside considerations of principle, we may notice that an individual (and therefore a people) that is educated to coercion is an individual who is educated to use his own knowledge less for his own purposes and more for the purposes imposed on him by others, often the majority, or the authority that represents it. From a quantitative point of view, it can easily be seen that peripheral knowledge (which, as we have seen, is of a better kind) is much more abundant than centralized knowledge, and therefore its use is much more likely to have successful results than the use of the latter. In fact, in our example of the roundabout and the traffic light, even if we gave the arbitrary value of 1 to every “unit” of knowledge used (that is, even if we did not take into consideration the superior quality of peripheral knowledge discussed above), assuming that that particular crossroad was crossed by 1000 cars a day, we would see that, in one day, in the case of the traffic light, the knowledge used would be 1 and in the case of the roundabout it would be 1.000. In one month, the amount of knowledge used in the case of the traffic light would still be 1, while that used in the case of the roundabout would be 30.000. In one year, the first would still be 1, while the second would be 365.000. If we introduced the qualitative difference discussed above and gave a unit of peripheral knowledge a higher value than the one given to a unit of centralized knowledge, obviously the quantitative gap between the two would be even larger. In conclusion, we can say that the quality and the quantity of knowledge used in the roundabout case are superior to those used in the traffic light case. This does not mean that the traffic light should never be used, of course: in extraordinary situations (e.g. at particular crossroads where roads are very large and traffic is massive and continuous) the benefits of arbitrariness (that is of the use of centralized knowledge) may be less than the relative costs. However, it is important to underline the fact that this is an extraordinary situation, that is that in ordinary circumstances traffic lights should be limited as much as possible. And it should also be reminded that, in this discussion, we are not considering matters of principle.
From the traffic to the economy
When we move from the “roundabout vs. traffic light” metaphor to the economy, the problems become obviously much more complex. Their structure, however, (and therefore the abstract criteria of assessment), remains almost the same. In the economy, a situation similar to the traffic light case (in which centralized knowledge is valued), is that of government intervention, where decisions tend to be strongly concentrated at central level (say, planning, nationalization or excessive/bad regulation). On the other hand, a situation similar to that of the roundabout case (in which peripheral knowledge is valued), is that of the so-called “free market”, or, better, catallaxy, where decisions tend to be left in the hands of those who are responsible for the relative actions and have knowledge of the particular circumstances of time and place. Finally, a situation similar to that of the extremely large roads with constant and intense traffic where traffic lights could be justified on grounds of expediency, is that of the so-called “market failures”, that is those extraordinary situations in which the free market cannot operate, such as the case of natural monopoly, for instance. It is important to emphasize that these situations are normally “extraordinary” because of their structure, not because of their consequences: a form of government intervention may be justified in the case of the railway between Naples and Milan, as there can be only one and competition would not be possible. On the other hand, the same intervention could not be justified to help an airline that is going bankrupt (that is, in a situation which is extraordinary for its consequences), for, in the case of the airline, competition can work and bankruptcy of inefficient companies is precisely the result of competition. In what follows, let us call arbitrary a measure (or a policy) which values centralized knowledge (or which values it, or uses it, more than it is strictly necessary); vice-versa, let us call non-arbitrary a measure or a policy which values peripheral knowledge (or which limits the use of centralized knowledge as much as possible). From what we have seen earlier, given the difference in the quantity and quality of the knowledge used in the two cases, we can say that, in normal circumstances, an arbitrary measure or policy will always tend to produce much worse results that a non-arbitrary one. In the case of the economy, however, the argument is more complex than this.
Individual differences
One important difference between the case of the economy and the “traffic light vs. roundabout” one is that, in the economy, the individual judgement applied in the use of peripheral knowledge depends not only on the particular circumstances of time and place (as it does in the roundabout case), but also on the individual characteristics of the person who has to make a decision. In other words, while in the roundabout case individual differences did not matter (in the sense that they were not able to significantly affect the decision whether to cross or not to cross), in the economic case individual characteristics matter a lot. In other words, when we move from the “roundabout vs. traffic light” metaphor to the economy, the case for the use of peripheral knowledge is even stronger. Let us consider the following example: because of a series of coincidences, Andrew and Claire may have a concrete business opportunity which they expect will generate a lot of cash but will require an extraordinary amount ofwork (say, also during week-ends and evenings). Because of his order of priorities (e.g., we can assume that Andrew gives a great importance to sailing during week-ends and spending the evening with his girlfriend), Andrew may be unwilling to take this opportunity. In other words, he prefers having significantly less cash and more free time, which, in this particular moment, he values a lot. Claire, on the other hand, in the same particular moment, does not have a boyfriend, nor, say, a passion for a sport that requires that she is away during week-ends; the gratification Andrew takes from sailing, she takes it from living in a beautiful house, with beautiful, expensive objects. Assuming (temporarily, and just for the sake of the argument) that individual utilities could be compared and even measured, and that Andrew's and Claire's utilities (after having taken their respective decisions) were the same, we can see why an arbitrary measure of the government transferring resources from Claire to Andrew (e.g. via progressive taxation) would be unfair. In fact, this measure would be based on centralized knowledge of a situation where centralized knowledge is irrelevant: by transferring resources from Claire to Andrew, the state is imposing a centralized view of priorities, while these depend exclusively on peripheral information, that is on information which is not available to the state. Of course, the cases of John who had an accident and who cannot have access to healthcare because of his limited financial resources, or of Catherine who cannot afford a roof, are different from the case of Andrew (who simply “has less” than Claire, but enough to have a decent life (1)) and in some circumstances they could be qualified as extraordinary and justify the use of arbitrariness, as long as this is limited (both in quantity and in form) to provide John and Cathrine with the “basic needs” they cannot afford (2). However, taxing Claire and Andrew (even via the arbitrary measure of progressive taxation) for providing John and Catherine with the basic needs they lack, is something entirely different from taxing Claire in order to transfer resources to Andrew (i.e., the middle class) just because someone believes that Andrew's tastes are more deserving (and, knowing that the “Andrews” are the majority, because their votes are needed to be elected).
Examples of arbitrary measures Even though it is one of the most common examples of the improper use of centralized knowledge, and of the coercive effects that this produces, progressive taxation is hardly the only one. In Italy, for instance, the length of any rent contract for residential property is decided by the state (eight years), and so is the use that one can make of his own property (say, whether one wants to use an existing space, without changing its structure, as a barn or as a house), and even the distribution of the space of his own flat, provided that all the alternatives are equally sustainable from a structural point of view; not to mention all the other areas (first of all that of the so-called “market” of labour) in which the space left to the use of peripheral knowledge (which implies freedom of contract) is reduced practically to zero. These are all cases where most, if not all, relevant information is of a peripheral kind, but in which the space for the use of peripheral knowledge is reduced to very little, the result being arbitrary coercion and loss of opportunity. Recourse to centralized knowledge beyond situations which are extraordinary because of their structure (and, only in extreme cases such as those of the lack of “basic needs”, because of their consequences) tends to increase when the recourse to peripheral knowledge would be more needed, that is in moments of crisis. These moments tend to be taken advantage of by authorities in order to extend their control over the economy and protect the short-term interests of those whose votes they need in order to be re-elected. In so doing, authorities reduce the amount and the quality of knowledge available for recovery and drastically reduce the opportunities on which every recovery from a crisis is necessarily based. In other words, by basing their action on arbitrariness, the authorities delay and sometimes impede a recovery from a crisis and destroy many of the conditions (not only economic) for it to happen.
Rational vs. spontaneous order So far we have spoken of single measures or policies, of how and why they can be arbitrary or non-arbitrary and of the effects that they tend to produce. These measures or policies are not only important in themselves, because of the particular results they produce, but also, and perhaps more importantly, because of the different orders that they beset. In order to have a visual image of the difference between designed orders produced by the use of centralized knowledge and the spontaneous orders produced by the use of peripheral knowledge, let us go back to our “traffic light vs. roundabout” metaphor. In particular, let us imagine to be on top of a skyscraper and to look down at a traffic light, on one side, and at a roundabout, on the other. On the one hand, we can see that the traffic light produces a kind of order, with cars moving and stopping at the same time. This order is based on design: in particular, knowing the timing of the traffic light, one can exactly predict whether in x hours, or days, or years from now, a car finding itself on one side of that crossroad can cross it or has to stop. On the other hand, we can see that the roundabout also produces an order, with cars crossing the roundabout when they can. As Hayek would say in the economic case, this order is the product of human action (cars crossing the roundabout) and of general, abstract rules or principles (cars have to give way to those who are already crossing the roundabout), but not of human design: in fact, it is impossible to predict whether in x hours from nowa car finding itself on one side of that crossroad can cross it or has to stop. Hayek calls this order a spontaneous order: an order which is the result of human action (i.e. of the spontaneous interaction between individuals), but not of human design. Now, the important point is that the absence of design (that is, the absence of arbitrariness, of coercion, of direct control, of use of centralized knowledge) does produce an order: in other words, it does not produce chaos. It produces an order, different from the first (i.e. better, for the reasons we have discussed above), but still an order. When we move from the “roundabout vs, traffic light” case to the economy, again, the situations are much more complex, but their structures remain essentially the same. Arbitrary policies (that is, policies which value the use of centralized knowledge) tend to produce a designed order, that is an order based on arbitrary commands and therefore on coercion. Vice versa, non-arbitrary policies (that is, policies which value the use of peripheral knowledge) tend to produce a spontaneous order, that is an order based on general, abstract rules and on freedom. Because of what we have discussed above, leaving aside considerations of principle and not taking into considerations organizations (e.g. what works in a business), spontaneous orders tend to be much more successful than designed orders because they make use of far better and far more knowledge. It is essentially on the base of this conclusion that Hayek successfully predicted the collapse of communism, which fortunately happened just in time for him to observe it. Why, then, imposing arbitrary measures?
Promotion of particular interests One obvious reason is particular interest: as we have seen, in countries where measures are not effectively limited by abstract principles, arbitrary measures allow policy-makers to protect the particular interests of those whose votes they need in order to be elected. Since the time horizon of both parties (policy makers and electors) is short (3), there is no real obstacle to the implementation of such unsuccessful and coercive arbitrary measures, at least until the problems generated by them risk to endanger the whole system.
Fear and inability There is, however, in my opinion, also another reason why arbitrary policies continue to be imposed. This reason is fear; a fear which is produced by the inability to conceive the existence of a spontaneous social order or, which is essentially the same, by the obsession of direct control. Some individuals simply cannot conceive an order which is not the result of human design. As Hayek says, for them the absence of coercion, the use of peripheral knowledge, is chaos: “Those (…) who cannot conceive of anything serving human purpose that has not been consciously designed are almost of necessity enemies of freedom. For them freedom means chaos” (Hayek, 1960, p.61). These enemies of freedom are not necessarily wicked, they just lack a particular ability, while perhaps having many others. In addition, it seems that those who, for one reason or another, lack this ability tend to be the majority: this is one of the reasons why so-called “democracies” (totalitarian democracies) tends always to be more and more a designed order. Whether this situation can be reversed or improved, or not, is not our concern here. The point I want to make is that basing a policy on non-arbitrary policies does not imply chaos, but rather a superior form of order; it means to rely on abstract, general rules (or principles) and renounce controlling the use that individuals, within those general, abstract principles, make of their freedom. As Hayek says, “It is because we do not know how individuals will use their freedom that it is so important” (Hayek, 1960, p. 31). In other words, it means to renounce to control the details of the order produced by the spontaneous interaction among individuals, but instead to focus only on the fundamental, general laws on which that order is based. It means to use peripheral knowledge as much as possible, and centralized knowledge as little as possible. When Newton, completing a scientific revolution which was started by Copernicus and taken further by Kepler, discovered the abstract laws that governed the movements of the solar system, he had to admit that he could not control the details of the movements of the particular orbits, because the shape of every individual orbit depended on particular circumstances of time and place (the gravitational interaction of every planet and of the sun with every other planet) which was impossible for any person to know: “By reason of the deviation of the Sun from the centre of gravity, the centripetal force does not always tend to that immobile centre, and hence the planets neither move exactly in ellipses nor revolve twice in the same orbit. There are as many orbits of a planet as it has revolutions, as in the motion of the Moon, and the orbit of any one planet depends on the combined motion of all the planets, not to mention the action of all these on each other. But to consider simultaneously all these causes of motion and to define these motions by exact laws admitting of easy calculation exceeds, if I am not mistaken, the force of any human mind” (Newton, De Motum Corporum in Gyrum). This parallelism with the history of astronomy could be taken very much further, which here I will not do for reasons of prolixity, more than for reasons of space. However, it is interesting to note that the scientific revolution transformed what, before it, was conceived as a designed order into a spontaneous order. In fact, in the Ptolemaic system (which was the dominant system, and as such based on Aristotelian cosmology, at least until Kepler, and which was definitely displaced by Newton discoveries), the shape of the orbits, their relative position and even their order was arbitrary, that is decided by (scientific, in this case) authority; and so were the reasons given to the apparent “retrogression” of the planets when seen from the earth. In Newton's system, on the contrary, the space for arbitrariness was zero (and this absoluteness is, perhaps, the main difference between natural sciences and social sciences): the shape, position and order of the orbits depended on particular circumstances of time and place and was governed by a few abstract rules, as was the apparent retrogression of the planets. While in the former case the laws could be “made”, in the latter they could only be “discovered”: this should remind us something about the difference between the totalitarian concept of law (in which the law is the command of the authority) and the liberal one (in which law exists before legislation and is, if necessary, discovered).
Conclusion In this paper I have defined the arbitrariness of a policy on the grounds of the kind of knowledge it values. In particular, I have defined arbitrary a policy which tends to value centralized knowledge, and non-arbitrary a policy which tends to value peripheral knowledge. I have then argued that, in the long run, non-arbitrary policies tend to produce much better results than arbitrary policies, and that these tend to be bound to failure. In my argument I did not take into consideration matters of principle: if these were taken into consideration, in many instances non-arbitrary policies should be adopted even if they were less effective than arbitrary ones. I have also argued that the superiority of non-arbitrary policies is not absolute, and that in extraordinary circumstances the reverse is possible. These circumstances, however, normally have to be extraordinary because of their structure, not because of their consequences. One exception in which arbitrary policies could be allowed in circumstances which are extraordinary because of their consequences, and not because of their structure, is that of the lack of “basic needs”, of which I gave a personal definition. Finally, on a more general level, I argued that while arbitrary measures tend to produce designed orders, non arbitrary measures tend to produce spontaneous orders and that not only are these more vital and luckily to produce better results, but they also imply much less coercion of some by others: in other words, unlike designed orders which tend to be based on slavery, spontaneous orders tend to be based on freedom.
> Share:
NOTES (1) The fact that Andrew's position, unlike John's or Catherine's, is the result of his choice, is irrelevant because the equality of utilities was artificially introduced as a simplifier of the argument, not as a “requirement of justice”, which it can easily be argued that it is not, even though the egalitarians obviously would not agree with this, though probably without arguing in favour of this assumption (see for example Parfit, 1997). The argument that material equality is not a requirement of justice is beyond the scope of this paper. (2) The concept of “basic need” is complex and subject to interpretation: its discussion is beyond the scoper of this paper. (3) In the case of the policy-maker because it is limited by the legislature, and in the case of the elector because his actions are guided by his private, often short-term, interest and do not take into consideration the system as a whole, like in the case of the airline who would be happy to receive financial support by the government, even of this would disrupt the competition which is good for the system as a whole.
REFERENCES Dworkin, R., 1977, Taking rights seriously (Duckworth, London). Dworkin, R., 1986, Law's empire (Fontana Press, London). Hayek, F. A., 1945, The use of knowledge in society. The American Economic Review 35, 519-530. Hayek, F. A., 1960, The constitution of liberty (Routledge and Kegan Paul, London). Hayek, F. A., 1982, Law, legislation and liberty (Routledge & Kegan Paul, London). Parfit, D., 1997, Equality and priority. Ratio, 202-221.
|
