General economic policy



01. Approaches

02. Methods

03. Target analysis

04.  Means analysis

05.  Promoter analysis

06. Political economy

07. Welfare economy

08. Order analysis

09. Order conception

10. Order dynamics


Chapter 4: Means analysis


0th Introduction

1th The question about the concretization of means

2th The question about the intrinsic value of means

3th The question about the efficiency of means

4th The question about negative secondary effects

5th The question about the market conformity of means

6th The political feasibility of means




0th Introduction


This section deals with the question of which contribution the economic policy teaching can make in connection with the question of what means shall be used to achieve a given economic policy aim. We have already pointed out that this question can by no means be answered solely by science, that the question of the appropriate means for a given aim can not be deduced only from factual connections, which can be clarified definitely on the part of science, that here always value judgments play a part, too, which can only be rendered by the politician who uses these means.


Now, what are the issues which the science can answer in the scope of the means analysis?


Firstly, it must be clarified in which the essential properties of the means up for discussion exist anyway (question of the concretization).


Secondly, the question arises whether the politicians (the public) grant a positive or a negative intrinsic value to the means up for discussion (question of the intrinsic value).


Thirdly, it is necessary to clarify whether and to what extent the means to be used are also able to achieve the desired effects (question of efficiency).


Fourthly, it is to check whether the intended use of means leads to undesired or even desired side effects at variables that should actually not be affected (question about the secondary effects).


Fifthly, it is to clarify whether the individual means are in conformity with regard to the existing economic system, that is whether, resulting from the existence of a market economy order, the means to be used has to be considered as market-compliant (question of market conformity).


Sixthly and finally, we must also reckon with the possibility that certain means recommended by science are not allowed due to the constitution of a state and that is just why they are not up for discussion. It is common knowledge that Keynesians recommend to combat cyclical unemployment thereby that the state expands its expenditures and finances these additional expenditures not by taxes, but with credits from the central bank. However, such a deficit in the state budget is not permitted in accordance with the Basic Constitutional Law since a while. Such measures (deficit-financed government expenditure increases) can therefore not be introduced in Germany.


Nevertheless, it must be allowed to demand fundamental law changes, which also allow such measures, as long as the scientist is convinced that these measures do not violate the fundamental values of our order and are therefore prohibited unjustly in the Basic Constitutional Law.



1th The question about the concretization of means


Let us begin with the question of the concretization of a means. Similarly, as in the definition of an economic policy aim also in the context of a means analysis it is to clarify as a first step, wherein the characteristics of the particular means consist. However, the concealment trend that we have seen in the context of the aim analysis is given less to the means. It is rather primarily about that means may have significant and insignificant qualities and that only such means appear suitable which possess the essential characteristics.


The question of the necessary properties can only be answered on the basis of a theory. A theory shows how determining factors can be traced back to an economic event. Within the scope of the policy, the issue is to bring about a specific, desired event and therefore this event becomes the aim. The means is then exactly appropriate to bring about this desired condition when its properties apply to the necessary and sufficient determinants.


Let us make these theoretical considerations clear to us in a practical example. Politicians would pursue the aim to participate workers in operations side acquisition of assets and to increase their total income in this way. Up to discussion would be the introduction of a participation wage. If the means of participation wage is chosen, then a part of the wages is saved compulsorily for the workers, those savings remain either in the company which applies this participation wages or are transferred to credit institutions.


Let us consider the characteristics of a participation wage more closely. A participation wage could be granted by a company without a contract, and thus without any commitment. This property is indeed of great importance for the question of which rights the worker receives, if he can possibly sue for the participation wage. For the question of whether an actually granted participation wage is, though, able to increase the total income of workers, this question is of minor importance, in this economic context it does not count to the essential characteristics of the participation wage.


Now, if we put the theory of distribution developed by Kaldor as a basis, we have to insinuate that an increase in the employee income is only achieved when the savings rate of the employees’ increases; so we need to verify whether a participation wage causes the saving rate of the employees to rise. The participation wage is not paid to the employee, but saved compulsorily, so one will able to insinuate that the savings rate of the employee is raised at least temporarily, and that therefore an essential prerequisite is created for that the participation wage leads to success. This means that the coercive character of the participation wage is one of the essential characteristics of the participation wage; it is the compulsion that triggers the desired effect.


However, it must be seen that an employee who has savings already, has the option to dissolve his previous savings; in this case, the savings rate of the employee does not rise or not to the full extent. It must also be taken into account that at each investive wage scheme the blocking period expires some day and the employee receives the right to dissolve the savings. Only if there are good reasons that employees do not entirely dissolve their savings after the blocking period, the savings rate of employees has increased in the long term and only then it can be expected that the investive wage achieves its purpose also in the long term.



2th The question about the intrinsic value of means


In and of itself, means are not introduced for its own sake, but in order to achieve indirectly other superordinate aims. In its capacity as means, means initially do not have any intrinsic value. Nevertheless, it must be expected that measures that are used as means are not only considered as means, but can also obtain an intrinsic value, namely a positive or negative value. Initially, let us make this relationship clear to us with an example from the private sphere.


An enterprise would pursue the primary aim to achieve the highest possible profit. The profit could in certain circumstances also be increased in that the said entrepreneur defrauds or extorts his customers or suppliers. Therefore the fraudulent behavior or the extortion would be in fact a means to increase the profit of the enterprise. Nevertheless, no one could recommend to an entrepreneur to resort to these means and one will sanction such behavior. Fraud and extortion are matters of fact which are classified as criminal and not even the highest material success can justify the use of these means. So we adjudge a high negative value to this kind of means.


If on the other hand an entrepreneur is willing to donate generously for the education of children from developing countries, so this donation may be discussed, among other, under the aspect that generous donations could raise the image of this entrepreneur and that in this indirect way even the sales volume could increase. Important is in this context, however, to note that this donation has not to be judged  only, and not even severely, from the view point whether this desired increase in sales volume is expected, one will see much more a positive intrinsic value in the donation as such.


These connections also apply to the use of political means. Again, we have to assume that certain measures as such are judged to be positive or negative, regardless of their effects. So the Keynesian theory recommends providing a deficit in the national budget in order to increase employment. The national debt is, however, considered by many people as something negative, regardless of whether it can be assumed that a national debt produces positive effects on employment.


Now, what is the contribution of economic policies teaching in this context? Of course, science is not capable to prove such a review as incorrect. It is the right of each citizen or politician to reject a particular activity - in our case the national debt - as such. Here, it is essential to consider that such assessments in turn have their specific causes. This negative assessment of government debt could derive therefrom that firstly one assumes that a debt of a private household is undesirable and that what is true of households, could neither be wrong for public budgets.


Here, science can demonstrate that this reasoning is based on incorrect conclusions. First, it can be shown that a debt of a private budget is only undesirable if the loans are provided only for consumption and when it can not be assumed that the budget could expect increases in income in the near future.


This consideration must not be simply transferred to the debt of public budgets. A significant difference between the debt of a private and a public budget is that the household incurs debts to another business entity and burdens this, while the state incurs debts to its citizens, thus effectively to himself.


One can add very well some points against a national debt; so price-raising effects may arise from a deficit, future generations may be affected if the government debt leads to case that the readiness to risky investment decreases.



3 The question about the efficiency of means


The central question within the means analysis refers to the suitability of economic policy means. When speaking of the contribution of science to the means problem, one generally thinks of the suitability question. The relationship between theory and politics results - as already indicated - from a socio-technical reformulation. The aim-means relationship corresponds to the cause-effect relationship, the aim is the intended effect, the means the set cause.


An efficiency analysis can here be approached from three sides:


Firstly, one may ask whether a given means is suitable for achieving a specific aim. Here aim and means are given, it is asked for the theoretical relationship between these two variables. For example, based on the theory of distribution shall be examined if the given aim: Wage ratio increase can be achieved by the introduction of the means: participation wage.


Secondly, one can ask for the characteristics that a means is required to have to allow that the given aim can also be realized. Here the aim and the relevant theory are considered to be given, it is asked for how a suitable means has to look like.


Thirdly, one could also ask the question, which aims could be achieved with a given instrument. Here the aim is the problem size to be examined; here the means and the theory to be applied are given.


Now, one can distinguish between an absolute and a comparative efficiency analysis. Within the scope of the absolute analysis one confines oneself to check whether a very specific instrument is able to achieve the aim. So could, for example, a representative of the neoclassical theory come to the conclusion that a reduction of the key interest rate of the central bank would be able to boost the economy.


Within the scope of a comparative efficiency analysis, several instruments are checked for effectiveness and it is investigated, which of the means under discussion comprises the highest efficiency. Let us take again the example of a planned economic recovery. To be available would be now on one side a reduction of the key interest rate of the central bank and on the other side a deficit-financed increase in government purchases.


Here, a Keynesian will suggest choosing the means of increase in government purchases. With an interest rate reduction there is the risk that private investors do not respond to the interest rate reduction as hoped with an increase in investment volumes. In times of recession the elasticity of investment demand would namely be low because on one side, due to the strong competition in product markets, interest rate reductions would have to be passed on in the goods price. And on the other side, since in times of recession all production capacities are underutilized in any case due to decline in consumer demand, the entrepreneurs are not interested in increasing the capacity any further. In contrast, the effect of a deficit-financed increase in government purchases would be sure, so that an expansionary fiscal policy would lead to success in any case.


Furthermore, one can distinguish between a qualitative and a quantitative efficiency analysis. At a qualitative analysis one confines oneself to check whether a positive effect on the aim variable has to be expected at all. At a quantitative analysis, however, it is also about how much the instrument variable has to be raised in order to achieve the desired effect.


However, such a quantitative analysis presupposes that both, the aim and the use of means can be quantified and that the underlying theory enables a quantifiable relationship. Let us bring again the example of the economic recovery. The aim consists in an intended increase of the domestic product of 12 billion. The use of means, the government expenditure increase, can also be measured in monetary units and the relied Keynesian theory of the multiplier permits the conclusion that, under the given conditions (i.e., for example, at a savings rate of 25%), an increase in government spending of 3 billion would be necessary for an aim realization.


Generally, it is assumed that within the scope of a policy which intends to influence the economic process, the quantifiability of aims and means is given, while under the regulatory policy one often has to be confined to a qualitative analysis.


Within the scope of the quantitative analysis also the problem of critical thresholds is broached. We can not expect that the relationship between the aim- and means variable is continuous and proportional, rather we have to assume that means often do not have an effect at all until a certain extent, because e.g. in the case of a too small extent, the addressed public does not yet take note of these measures. So may minimal interest rate reductions fizzle out in their effects.


On the other hand it has to be reckoned that at some instruments from a critical height on the effect turns into its opposite; so a duty rate increase will lead to additional receipts in customs duties only up to a critical limit. Duties are namely added usually to the product price, but if the elasticity of demand is greater than one, then the price increase will be overcompensated by the reduction in quantity, the sales and with it the customs receipts then go back despite duty rate increase and price increase.


Finally, a third distinction is of importance: One can distinguish between static and dynamic efficiency analysis. At the static analysis we confine ourselves to how the equilibrium variables in the aim variable change due to the use of means. An increase in government expenditures by € 1 billion would lead after the attainment of a new equilibrium to an increase of the domestic product of say € 3 billion.


If one expands, however, the consideration to a dynamic analysis, so one will also examine which time it takes for a used means at the point in time (t1) to have an effect on the aim variable. Such an analysis requires that projections about the course of the aim size are possible and, moreover, that it is known how long the to be examined effect process takes between means application and aim size. If we can assume e.g. that a government expenditure increase will affect the level of income only within 1 1/2 years, so this knowledge can only be used if one can prognosticate how the income will change within 1 1/2 years. Because only in this case the means application can be dosed properly.


The efficiency analysis is then faced with difficulties when measures are under consideration, which are introduced for the first time. Then there is a lack of empirical values about which influence is actually emanating from the newly applied instrument. So at the first launch of the participation wage it was unknown how the employees will behave when the blocking period expires and the employees have the opportunity to withdraw the compulsorily saved funds from savings accounts. Just because no experiences existed in this matter, there were different hypotheses about the success of these measures.


A second difficulty arises from the circumstance that at almost all the political measures a part of the population will be disadvantaged, and that this part of the population will strive to avoid these disadvantages. A rational policy will need to include these potential avoidance mechanisms in their consideration. Here, it must be expected that the policy will always lag behind the development, because the burdened groups are constantly developing new behaviors in order to escape this burden.


If e.g. the state tries to slow down the consumer demand in the framework of a Keynesian policy of fighting inflation by tax increases, and if the trade unions strive to compensate the by the tax increase expected reduction in the net wage income through increases in the gross wage rates, then just by this behavior of trade unions the efficiency of tax increases will be reduced. If the state nevertheless strives to remain successful in fighting inflation, it requires flanking measures to avert this danger.


Such a flanking measure could e.g. consist in that the state uses financial incentives for the introduction of a participation wage. If now the trade unions try, due to these government incentives, to avert the reduction of their income by means of financial statements of participation wage contracts, the policy remains successful to fight inflation, at least when we assume, following the demand-driven inflation theory, that participation wages do not act price increasing.


Furthermore, flanking measures can become necessary also for a second reason. In general, we must in fact assume that the problem sizes of economic theory depend on several determinants and that is why several means must be used in order to get successful. Let us take again the example of the economic policy.


According to the Keynesian idea the attempt will be made to revive the economy by a deficit-financed increase in government expenditures. Now it is feared on the part of neoclassicists that the additional supply of securities on the capital market exerted by the state leads to interest rate increases, and that therefore private investments decrease and that thus the success of the economic revival will be nullified or at least reduced. To prevent this crowding out, the Keynesian economic policy now tries to persuade the central bank to an expansion of the money supply; in this case, the interest rate increases are reversed, so that the feared crowding out holds off.



4 The question about negative secondary effects


Political measures generally do not only affect the variables that are to be influenced by this measure. We rather have to assume that as a rule other variables are influenced positively or negatively. If it is a matter of positive effects, we speak of welcome side effects or positive secondary effects, but if other variables are tangent negatively, then there are negative secondary effects on hand.


Positive secondary effects are e.g. at hand, when growth policy measures do not only affect the economic growth positively as desired, but do also affect the employment. These effects are unproblematic; we thus do not want to deal any longer with these effects subsequently.


Negative secondary effects are at hand, however, e.g. if an expansive fiscal policy (deficit-financed increase of government expenditure) leads not only to the intended employment increases, but at the same time effects negatively - unintentionally, perhaps not even expected - on the inflation rate. This type of secondary effects brings a number of problems along with which we want to deal in more detail in the following.


Negative secondary effects are closely linked to the problem of conflicting aims. One and the same connection exists when we speak of means related trade-offs or of negative secondary effects. The adverse impact of an employment policy to the inflation rate, for example, means that the aim of full employment is in conflict with the aim of monetary stability.


In a much greater extent than it is true for the efficiency analysis, we have to assume that for the effects which are classified as secondary effects there is no sufficient theoretical knowledge at hand, so that it can only be speculated about possible secondary effects, so that often the presence of certain secondary effects is known only much later.


To this, let us first consider the applied procedure in the context of efficiency analysis. It is the concern of a theory to inform over preferably all determinants of a problem size. So the Keynesian inflation theory assumes that only the increase in the effective demand is responsible for goods price increases; as we subdivide the total demand in consumption, investment, government and finally export expenditures, we have included with this list all known possible determinants of inflation (in opinion of the Keynesians).


This target completeness in the number of determinants is generally also possible, because we can assume that this connection between determinants and problem variables takes place in a relatively short, reasonable time and can therefore also be detected at an accurate empirical observation, generally.


Other applies generally to the occurrence of secondary effects. Here one usually assumes that prolonged periods elapse before the undesirable secondary effects of a measure occur. Let us take again the example of a Keynesian employment policy. The theory teaches us that positive employment effects can already be detected very soon after increase in government spending. Finally, government expenditures which are used for the purchase of goods and services pose already income growth.


In the context of the inflation theory it is, however, pointed out that these state expenditure increases generally also lead to price increases, but that the time lag situated between expenditure growth and inflation is substantially greater than the time lag between expenditure increase and employment growth.


This is also the reason for the popularity of an expansionary employment policy among politicians. If they start with these measures shortly before the election, they can hope that the employment increases will still have a positive effect on the election, but that the undesirable price increases occur only after the election, and will therefore not influence voter behavior. Since it can be assumed that the voters forget relatively quickly and their voting behavior depends solely on the events just before the election, the inflation effects will not be counted against the politicians also in the long run.


Just this prolonged period of occurrence of secondary effects now brings about that within the framework of the efficiency analysis we may have theories that list preferably all of the known determinants, but that there is no theory that summarizes all the possible secondary effects. If we inquire about the secondary effects of an expansionary employment policy, we learn about the to be expected price increases of the inflation theory, about the growth effects of the growth theory, about the adverse distribution effects of the distribution theory and about the possible misallocation of the allocation theory.


These different procedures within the framework of the efficiency- and secondary effect analysis now bring along that the progress in the development of theories in connection with the secondary effects is much lower than in connection with efficiency problems. If namely the prognosticated effects of a theory do not occur or obviously other determinants of the occurrence of a problem size are accountable, so there is also a strong incentive to modify the theory. In the context of research the necessity results to modify a theory until finally all identifiable determinants of a problem size are included in the theory.


The process of knowledge discovery takes place quite differently in the context of secondary effects. There is no method of falsification, if we have assumed so far the false hypothesis that an expansionary employment policy affects only the inflation rate but does not have a negative effect on the allocation. It is not a single theory, but only parts of quite different theories that are on trial, so it is missing here the incentive to check, whether not also secondary effects on other problem sizes are to be expected. It is more a matter of chance that one comes across secondary effects and this applies all the more the less such connections are suspected and the longer the period of time elapses until these secondary effects occur. For these reasons may the analysis of secondary effects be to a greater extent of a speculative nature than the efficiency analysis.


Just for these reasons, on introducing new measures negative secondary effects are often prognosticated, which then have not occurred at all. And on the other side negative secondary effects of major extent have been overlooked which are only found in hindsight, perhaps actually not until many decades after the introduction of a measure.


The straightening of rivers, e.g. the Rhine, was celebrated as a great success in the history of Germany, because rivers became navigable in this way. Waterways emerged in this way, the travel costs have been reduced dramatically, and thus the productivity of the economy could have been substantially increased ultimately. Only much later - and in fact not until around a hundred years later - the negative consequences of this straightening became visible, which were reflected in multiplied and stronger inundations and high economic damage costing billions.


If one wants, one can count the financial costs of certain measures also to the negative secondary effects. Economic policy measures often entail further measures in the future, with the result that the financial costs associated with this instrument are much higher than initially estimated. Three cases can be distinguished here:


The extent of the incurring costs depends on variables which extent increases over time and therefore increases without modification of the law. We take the example of the savings premium law of the 60s of the last century, which provided that certain premiums were granted for long-term savings. At the introduction of this measure only a relatively small part of the population could make use of this law, with the consequence that the financial burden of the state due to this law was initially very low.


In the following period, however, more and more people came to enjoy this law because of the general increase in income, so that the financial burden increased to an extent that was not foreseen in this scope. This connection can now become a problem if on the one hand such a law has been launched just because the financial burden was low and if on the other hand it is politically difficult to retract a measure again which has been carried out already for a long time.


A second type of financial follow-up costs exists when the extension of the measure gets necessary out of the inner logic. So the introduction of a dynamic old age pension in 1957 was, among other things, based on that also transfer income recipients should receive a compensation for price increases and a fair participation of the economic growth. The inner logic of the pension reform of 1957 now brings along that other pensions such as e.g. injury pensions should also be dynamized. Again, the introduction of a law led in the long run to much more cost increases than initially assumed.


Thirdly, financial follow-up costs can eventually occur also due to certain political mechanisms. The fact that the state grants conservation subsidies for crises occurrences in certain economic sectors can call other stakeholders from other economic sectors on the scene and they can also ask for subsidies, arguing that equal rights have to apply for all; if one part of the population is helped in times of crisis, this aid has to be granted on the same terms to all economic sectors. Thus, the volume of subsidies can increase dramatically.



5 The question about the market conformity of means


Economic measures may be checked continually as to how far they are compliant with the economic system in which they are used. Walter Eucken and Wilhelm Röpke have formulated the criterion of market conformity for a market economical system. Thereafter only such measures can be considered as compliant with the market order which do not intervene directly in the market process and only change in an indirect way the data which influence the economic decisions.


Among the basic economic decisions are counted the determination of supply and demand of goods quantities, their prices, the production technology as well as the location of a production. A economic policy measure will be considered as market conform as long as these decisions remain among private households and enterprises; In fact they can influence the decision of the private, but the private business entity has to decide as before whether it maintains the supply or demand after the changed situation.


If the state introduces e.g. a sales tax, it contributes to an increase in costs, and this cost increase will usually lead to a change in the supply. Such data change exists also if the state excludes certain alternatives. So could a company increase their profits by fraudulent practices or by a monopolistic merger with the other market partners, but the state prohibits these practices. Not the prohibition of certain activities is hereby non-compliant to the market; as long as the individual is not forced to a specific measure, as long as several alternatives remain between which the individual is still able to decide freely.


K. C. Thalheim has now criticized this criterion because it would emphasize to much the qualitative side and would neglect the quantitative aspects. He instead proposes to distinguish between system necessary, system conducive, system neutral, system damaging and system destructive measures. A monopolization of the bank notes creation was system necessary, competition of enterprises was system conducive, the imposition of a sales tax was system neutral, the introduction of prohibitive tolls was system damaging, galloping inflation has to be regarded eventually as system destructive.


B. Steinmann has also criticized the fact that the market conformity is measured at Eucken only on the question, to what extent it is intervened in the market process. However, Walter Eucken formulated seven constituent principles of a market economical order and the market conformity had to be judged by the extent to which each of these principles get attention. Although the introduction of a prohibitive duty was no immediate intervention into the price process, but it violated the principle of openness of the markets and therefore had to be classified also as non-compliant to the market.


Finally, Theodor Pütz has attempted to develop the criterion of the market conformity further. Whether a particular measure may be considered as market-conform would also depend on the surrounding circumstances. So although national definitions of prices would be classified as non-compliant to the market under normal conditions, but the same interventions could help to stop a market economy threatening deflationary spiral of price and wage reductions and be so far considered as market compliant.


Secondly, it would also depend on the area in which a measure is applied, monopolies are generally market damaging, on the issue of banknotes, however, they are indispensable for the functioning of the market economy.


Thirdly and finally, also the extent of a measure would decide about the conformity, slight expansions of the money supply may promote the process of economic recovery; however, a doubling of the money supply in a short time would ruin the economy in all likelihood.


The criterion of the market conformity emphasizes certain characteristics that generally can be checked relatively easy. It is a question of classificatory assignment. In contrast, at the criterion of secondary effects the effect connections were emphasized which are not always openly obvious and can only be checked with the help of a theory.


Now we have to be clear on that even the conformity criterion is often associated with the idea that market non-conform measures entail unwanted effects and endanger the stability of the market economy. So Walter Eucken has advanced the hypothesis that planned economy measures were not limited to individual markets, but rather entail more planned economy measures in other areas. With the consequence that this process eventually would inevitably end in a total command economy.


One can now doubt whether combined systems are in fact unstable and necessarily end in a total command economy, eventually the post-war economy in Germany has started as a combined system and the planned economy elements have been removed gradually. Nevertheless, the experience with economic planning interventions shows that a limitation of those measures to a single market is not possible. So leads e.g. an artificial shortage on a market to a shortage on all downstream markets. Here, complementarity and substitutability decide how strong the expansion of individual market conformities will be in detail.



6 The political feasibility of means


In the chapter about the aims of economic policy we had investigated the question of the realism of individual aims. Similarly, with regard to political means the question can be asked whether these means can be used politically at all. The constitutions determine which policy measures are allowed and which are not allowed.