10th Keynesianism




1st Introduction to the problem

2nd The main representatives and opponents

3rd Starting point: Say's law

4th Keynes' criticism on the Say's law

5th A simple C-I model

6th Criticism on the purchasing power parity theory

7th The alternative S-I model

8th Supply and demand for money

9th The Hick's IS-LM model

10th Integration of the employment function

11th Criticism on Keynes' assumptions 



10th Integration of the employment function



Until now we have regarded the (real) national income as an indicator of employment fluctuations. The last step is to show which relations exist between real national income and employment of workers.


Here we assume a macroeconomic production function, as this indicates how changes in the production factor input will affect the real domestic product. Here we will assume the Cobb-Douglas production function, which has been tested empirically on several occasions. According to this, the following correlation between domestic product (X) and employment (A) applies:

X = b * Aα * K1-α


Thereby, b represents a growth factor and α the production elasticity of labour (how many production units are produced additionally when the labour force is increased by one per cent). We want to consider this new function in our graph by drawing it in a new quadrant, the southeast quadrant. We plot the income level on the abscissa of the SE quadrant, but the number of labour units (labour hour or number of workers employed) on the abscissa downwards.


The employment function plotted in red assigns a certain level of employment to each level of real domestic product, whereby employment also increases along with domestic product. However, the circumstance that the law of the diminishing partial marginal product applies means that for each further increase in the domestic product, an ever-increasing quantity of employment per unit of goods is required. If we know the parameters of this production function, the growth factor b and the production elasticity α, we can read from this function how much employment will increase due to the growth in domestic product.



11th Criticism on Keynes' assumptions 


We will now turn to the criticisms put forward by the opponents of Keynesianism.


Underemployment as a special case

The first thing to note is that the Keynesian model in the version of the IS-LM model introduced by Hicks suggests that a functioning market process is quite capable of reducing underemployment. For this, let us look again at the diagram shown last.  


Let us assume that we are in a condition that corresponds to the domestic product E1, but that full employment could only be achieved with labour input A2 and thus at a domestic product E2. With the course of the IS and LM functions assumed in this graph, the state of full employment could very well be reached by either increasing the nominal money supply or by a general price reduction, which would also lead to an expansion of the real money supply. At the intersection of the dashed LM curve with the IS curve there is an equilibrium at full employment.



We now want to ask ourselves what course the two IS and LM curves would have to take so that Keynes' scepticism is still justifiable that the market would not find its way to a state of full employment by itself. Let us look at the newly drawn IS-LM diagram.



Here, the IS curve has such a low gradient that the intersection of this curve with the abscissa is at an income that does not guarantee full employment. Thus, in this example there is no positive interest rate at which full employment would have been achieved. The reason for this is, of course, an extremely low elasticity of investment demand regarding the interest rate. But it also matters how high the propensity to save is. If the propensity to save is high, a given increase in investment will only lead to a very small income increase. If, on the other hand, the propensity to save is extremely low, even small increases in the investment volume can trigger high income gains.


Secondly, the Keynesian apocalypse could also occur if the liquidity preference would be extremely high (almost infinitely high). For this, let us look at the next graph:



There, the LM curve runs (almost) parallel to the x-axis, it is assumed here that every increase in the money supply, no matter how large, immediately disappears in the speculative funds. In this case, the money supply in circulation remains almost constant, with the result that despite an increase in the money supply, the LM curve remains unchanged, it shifts to the right, so to speak. This means, however, that monetary policy is no longer able to shift the intersection of the IS curve with the LM curve in the direction of higher domestic product.


Now Keynes himself once expressed that this is a theoretically highly interesting case, but one that is very unlikely. Then it essentially remains Keynes' thesis that interest rate cuts in times of depression are not capable of bringing about full employment without the intervention of the state (thesis of the interest rate inelasticity of private investment).   


But precisely this assumption is extremely questionable when applied to the overall economic investment volume. In reality, two major limitations have to be made, even if one admits that for a certain part of the investments, interest rate changes indeed play only a minor role.


The role of rationalisation investments

On the one hand, Keynes' considerations on the justification for an interest rate inelasticity of investment are only valid for expansion investments. In fact, if enterprises have surplus capacities due to a general decline in sales, they will hardly be willing to continue to expand their capacities.


But there are also rationalisation investments that primarily do not serve to expand capacity, but rather to increase competitiveness based on cost reductions and quality improvements. And this motive occurs precisely in times when an entrepreneur is losing sales.


Now, a rationalisation investment could however cause new unemployment in a different way. Rationalisation often consists of saving production factors. If less labour is needed for production, this means that workers are laid off.


But whether this leads on balance to an increase in unemployment depends on a variety of factors. First, a part of the rationalisation investments consists of quality improvements. Second, cost reductions may not only result in reductions in the labour force.  


Thirdly, Streißler has shown in a historical empirical study that in the past there was much more capital-saving progress than labour-saving progress. Fourthly, even in the case of labour-saving progress, in addition to the redundancy effect in the industry in which these rationalisations take place, there is a compensation effect in the industries in which these machines are produced, so that at least a large part of the dismissals is compensated by new hires elsewhere.  


Fifthly and finally, the question of what kind of technical progress is applied depends crucially on the wage-interest ratio. If this wage-interest ratio corresponds to the real scarcity ratio of labour to capital, a technique will also be chosen that does not lead to greater underemployment. If, however, under a Keynesian policy, both a low-money policy (low interest) is pursued and, at the same time, wage increases that exceed productivity growth are enforced, then indeed a wrong wage-interest ratio contributes to generating unemployment.


The role of construction investment

The thesis of the interest rate inelasticity of investments is questionable in its generalisation also for a second reason. A considerable part of the investments is made in buildings. These investments have an extremely long production period. While machines are sometimes written off in three to seven years, investments in buildings and roads take many decades. As much as it may be true that interest costs do not play a decisive role in the context of total costs for investments lasting a few years, it can be assumed that for investments lasting almost 100 years even the smallest changes in interest rates lead to a considerable increase in costs, so that the investment behaviour with regard to these investments depends even to a large extent on the interest rate.


The importance of supply factors

In our previous considerations, we tacitly assumed that macroeconomic unemployment was in any case based on a lack of demand for goods. It is precisely this assumption that must be doubted. Especially in recent decades, supply factors have been responsible repeatedly for the emergence of unemployment. For example, increases in the costs of raw materials can lead to a radical change in the economic cycle and in this way trigger mass unemployment. Thus, in particular the two oil crises in the mid-1970s and early 1980s triggered a severe economic crisis and thus also mass unemployment.


Educational deficiencies can also be responsible for unemployment. On the one hand, an ever-increasing proportion of employees do not find employment because they do not have even the slightest knowledge. This is especially true for workers with a migration background.


But also, a shortage of highly educated skilled workers can trigger unemployment on a larger scale. Production requires almost always a number of highly educated skilled workers. If there is a shortage of these, other workers can also be affected by unemployment, since in this case production cannot even be started.


Relocation of work abroad

Furthermore, it must be considered that even if Keynesian policy in fact succeeds in triggering an increase in production through an increase in demand, it is by no means certain that this increase in demand for goods will also trigger a demand for domestic labour force. Enterprises are increasingly making use of the possibility of relocating parts of their production abroad since labour costs are lower there.


This example clearly shows that it is not enough to stimulate the demand for goods. At the same time, these examples draw attention to the fact that the wage level or the total labour costs do indeed help to determine the amount of work. It was the serious mistake of Keynesian policy to assume that prices cannot have any significant influence on employment.


The importance of time lags

Milton Friedman has pointed out that in the realisation of political objectives, a relatively long period of time elapses between the time when the problem arises and the time when the political measures take effect. In the context of political science, a distinction is made here between an inside lag and an outside lag. The inside lag refers to the decisions made by politicians, while the outside lag is the time needed by those affected by a measure to react to it and to make the necessary arrangements. This time span is subdivided into a recognition lag, the period of time that elapses before a problem is even acknowledged, a decision lag, which refers to the duration of the decision-making process, and finally a realisation lag, the period of time required to implement the decisions.   


Milton Friedman has pointed out that fiscal policy measures take about 1 1/2 years from the emergence of the problem, e.g. mass unemployment, through the passing of the laws or measures to combat this unemployment, until the increase in employment triggered by these measures.


This presence of a very long time lag now reduces the chances of success of these measures. In order to pursue a rational employment policy, employment policy measures would have to be introduced, for example, at a time when the economy is booming, and contractionary measures would have to be started at a time when mass unemployment is still prevalent. However, such an approach would be too much for politicians who are permanently under pressure to publicly justify their measures.


Adjustment processes

Milton Friedman also pointed out that trade unions and enterprises adjust to fiscal policy measures and question the success of these measures just because of their reaction. Thus, within the framework of a Keynesian employment policy, it is hoped that success will be achieved by the fact that the government's expenditure increases revenues and with them also the profits, and that enterprises will take these profit increases as an opportunity to increase production and thereby employment. Sooner or later, however, the trade unions will demand and enforce wage increases in the next collective bargaining negotiations as compensation for the price increases. This, however, will reduce profits to their previous level and the entrepreneurs will no longer have any reason to maintain the increased production.


Decisive is now that when entrepreneurs have realised that these increases in profits will only last for a short time, they are no longer willing to increase production and thereby employment, since it is difficult to dismiss workers once they have been hired, due to rigorous employment protection legislation, when production declines again. The success of an expansionary fiscal policy is then limited to the period in which the measures come as a surprise to those involved. As soon as enterprises and trade unions have taken note of these measures and adapted to them, the success of an expansionary employment policy disappears.


Political economic cycles

William D. Nordhaus has argued that Keynesian fiscal policy in a representative democracy can easily lead to even amplifying the cyclical fluctuations of the economy. Shortly before elections, politicians in power try to win votes by means of short-term employment programmes. Since these measures lead relatively quickly to a revival of the national economy, the ruling politicians can win votes this way and with them the elections. But after they have been re-elected, the politicians feel compelled to introduce contractionary, stabilising measures.


The employment programmes immediately before the election contributed to the fact that in the long run the increase in the money supply led to inflationary effects, which now had to be reduced by contractionary measures after the election. The forgetfulness of the electorate makes it possible for politicians to pursue such stabilising policies immediately after the election, since at the next election the electorate will have already forgotten these adverse effects shortly after the election. In other words, the politicians' attempt to pursue employment policy contributes to an ebb and flow between expansionary and contractionary policy and in this way amplifies the cyclical swings that should actually have been dampened by this very policy.


The importance of side effects

Keynesian policy is often judged unilaterally only by the extent to which it succeeds in increasing employment. However, it must be remembered that almost all policy measures have undesirable side effects on other policy objectives. Therefore, to assess Keynesian policies, it is necessary to examine all possible side effects.


Side effects are to be feared, firstly, on the goal of monetary stability. The expansionary measures of Keynesian employment policy are usually accompanied by a strong increase in the money supply. This increase in money usually has an inflationary effect only after a certain time. Initially, a large part of the money is kept in cash, which is only returned to the economic cycle when the first signs of an upswing become visible. If, during this phase, the central bank fails to withdraw the surplus money again, considerable price increases occur.


Economic growth and allocation can also be impaired due to an expansionary employment policy. This is especially true if government spending is mainly used for consumption purposes. Moreover, government programmes are often allocated by circumventing market-based efficiency controls, so that the production of goods does not correspond to scarcity conditions.


If the expansionary policy leads to price increases, these measures are also reflected in a passivation of the current account balance, especially if the foreign country adheres to a strict stability policy. If the prices of the goods to be exported rise disproportionately high, the export possibilities deteriorate with the result that the current account balance becomes passive.


Depending on the employment programme, the functional distribution of income can also lead to an undesirable redistribution of income in favour of enterprises. Most employment programmes provide subsidies to enterprises to give them an incentive to increased investment.


In public, the fear is expressed currently that a policy of indebtedness would lead to a burden on the future generation and would therefore be undesirable. If politicians get into debt today, the future generation had to pay back these debts one day and would be burdened in this way.


In reality, the answer to this question depends solely on the extent to which economic growth is impaired or even promoted by these measures. If investments in the future are channelled into education, for example, they will benefit the future generation and even deficit financing of these investments would benefit rather than burden the future population. Consumptive expenditures, on the other hand, burden the future generation if they are deficit-financed, since in this case resources are diverted from otherwise productive investments and are lost in today's consumption.


Finally, a deficit-financed employment policy can damage the political control mechanism of a representative democracy. Only with a balanced budget the voter is free to decide whether tax-financed government spending brings him higher welfare than if he had used these funds for private consumption goods.


If a politician can finance his expenditures in deficit, then voters only see the positive effects of government spending. Although the voter must also pay for these expenditures in the end, namely in the form of price increases, it remains unclear to him whether the benefit of government spending is really higher than the loss of benefit that results from the inflationary effects.