Chapter 9: The means of financial policy
01st Neutral financial policy versus fiscal policy
02nd Is a reflation sufficient?
03rd Built-in flexibility versus autonomous economic policy
04th Anticyclical versus compensatory financial policy
05th Deficit (without) spending, spending without deficit
06th The different financing methods of the deficit
07th About the problem of public debt
01st Neutral financial policy versus fiscal policy
Let us begin with the concept of a neutral financial policy. The starting point is the thesis of the self-healing power of the market economy. Therefore, the market economy does not need state intervention to eliminate imbalances; the market itself leads to optimal results. If a market imbalance occurs due to data changes, the market forces themselves ensure that this imbalance is reduced relatively quickly and without any intervention the state.
If there appears e.g. an excess supply on the labour markets which is manifested in unemployment, then wage reductions occur, and these wage reductions lead on the one hand to an increased demand for labour force on the part of the entrepreneurs and on the other hand to a reduced labour supply. Thus, the excess supply is reduced automatically from two sides.
According to this concept, the state achieves the neutrality towards market events precisely when its budget is balanced. A balanced budget is therefore regarded as a neutral budget! This is explained by the fact that the expansionary effects of the increase in government expenditure correspond exactly to the contractive effects of the increase in tax revenues, which means that both effects completely compensate each other.
Against these considerations, two points of criticism can be formulated above all. The first to be mentioned is the Haavelmoo theorem. Haavelmo has shown that, even with a balanced budget, the total demand for goods is affected by expansionary effects. Thus, in other words, the contractive effects of tax revenues do by no means compensate adequately the expansionary effects of government spending.
On the contrary, Haavelmo assumes that a simultaneous increase in government spending and tax revenues of €1 billion will lead to an increase in total demand for goods of just €1 billion. In this context, it is referred to that the multiplier of an expansion of a balanced budget is precisely the value one.
This thesis can be illustrated in two ways, in a static or in a dynamic analysis. In a static analysis, the total expenditure of an economy is divided into the expenses incurred by private households or enterprises and the remaining expenses incurred by the government.
Haavelmo now assumes with Keynes that the propensity to spend (propensity to consume) of households is always less than one, thus an increase in privately disposable income always leads to an increase in consumer spending. But that this increase is always less than the income gain, thus in other words a certain part of the income gains is saved and therefore the expenditures are no longer increased.
A certain restriction must be made, however. There is something like a minimum consumption amount that each household will exercise in order to survive at all. As long as the private disposable income has not yet reached this minimum level of spending (the subsistence minimum), there will either no savings be made at all or - if possible - more than the income level is consumed. In other words, the thesis that the propensity to consume of households is less than one only applies only from an income on that is higher than the subsistence minimum.
If we can still assume that the state will always ensure a balanced budget, then ex definitione the propensity to spend of the state equals just one. Thus, the government has a greater propensity to spend (=1) than the private sector (<1) has. And this means that whenever the state expands its budget and the tax revenues increase as much as public expenditure, there is not only an increase in public expenditure, but also the share of the state in the total (public and private) consumptive expenditures of an economy increases at the same time.
If, however, the share of the state in total demand rises, then by an income gain of one (e.g. one billion) the hereby triggered demand rises more strongly than before, since it is assumed that the government uses a larger percentage of its additional income for consumptive expenditures than private households. This means that the average spending propensity of the national economy will also rise if the state increases its budget balanced.
Now, we try to explain the statement of the Haavelmo theorem within the framework of a dynamic analysis. We speak of a dynamic analysis whenever we do not limit ourselves to the question, as in the context of a static analysis, of which variables and to what extent they are related, but furthermore try to clarify also how much time elapses in order that a variable x changes another variable.
In this context, the consumption function plays a decisive role. The consumption function indicates how changes in income affect consumption. Keynes limited the formulation of this function to the static statement that an increase in income triggers an increase in consumption that is lower than the variation of income, while the consumption function named after Robertson assumes that changes in privately disposable income in period 0 only affects in the next period 1, thus with a time lag in a variation of consumer demand:
Ct = f(Eprvt-1)
It is easy to explain that a certain amount of time (a time lag) elapses between the variation in income and the change in consumption until the income change affects consumption. For individuals to adapt to the changed situation, time passes, they must first become aware of the change in income, they must think about for what needs they will use an increase in income, perhaps the providers will finally have a delivery period.
Such a delay is understandable primarily because this change does generally not relate to essential goods. If a reduction in income is under discussion, it must also be considered that the purchase of some goods and services must be cancelled and that certain notice periods must be observed for this cancellation.
We now want to assume that in period 1 the privately disposable income (Eprv) was reduced by one unit because the state had increased the income tax rate. For this reason, only private disposable income is reduced in this first period, and not gross income. As expected, the change in consumption caused by this will only be reflected in the next period 2.
However, since the state had assumingly also increased state expenditure by the same amount in period 1, the income also increased by this amount at the same time. Here, the increase in expenditure may consist in the fact that the state raises the salaries of civil servants or purchases more goods, in this case the proceeds of the enterprises increase, and this increase is ultimately reflected in increases in profits or other incomes. Thus in total, the gross income increased in this first period by the increase in government expenditure.
In all subsequent periods, the domestic product remains constant, since the reduction in consumer spending caused by tax increases is compensated just by the consumption increases that occur due to the induced income increases. This shows that after the ending of this process which the expansion of the state budget had brought about, gross income had increased by just the amount of the increase in state expenditure. Thus, the multiplier effect is precisely one.
Now, let us turn to the criticism of the Keynes School. The Keynesians not only deny that a balanced state budget has no influence on national income, but furthermore deny the demand that the state should have no influence on the economy, i.e. behave neutrally regarding to national income.
In contrast to the Classicists, Keynes questions the Say's theorem, according to which every supply creates its demand by itself, and according to which macroeconomic unemployment can thus never be attributed to a lack of demand for goods. As already shown above, Keynes assumes that the capital market would fail, some of the savings were hoarded, and furthermore, in case of an increase in savings, interest rate cuts would not be able to expand the investment to such an extent that the employees, who had to be made redundant in the consumer goods sector due to the decline in consumption, could now be employed in the investment sector due to the increase in investments.
These deficiencies in the capital market thus led to a reduction in employment. Due to these deficiencies of the capital market, the state has the task of supporting the economy with financial policy means.
However, the classical counter-criticism is: Financial policy measures are inefficient because tax changes can only be implemented in the long term. Tax rates must be passed by parliament, and it takes a long time to pass tax laws generally.
A certain time period passes before the politicians even learn about the problems that need to be solved and have brought themselves to a proposal to solve these problems. This is followed by a first draft, which the cabinet then submits to parliament. This draft will be discussed in three readings in parliament. After the adoption of a corresponding law, the Federal Council usually must approve this law as well, since mostly the interests of the states are affected.
Due to the different compositions of the German parliament and the state parliaments, the ideas of the Federal Council often deviate from the legislative proposals of the German parliament. Here, a compromise must be reached in a time-consuming mediation committee of both bodies. If it is finally reached that the law has cleared all hurdles, it must be put into force by the Federal President. Furthermore, the application of this law requires numerous implementing rules. It has been estimated that from the first appearance of the problem until the implementation of these political measures, a time of about 1 ½ years will pass.
This means, of course, that these laws can only be successful if political activity is started correspondingly earlier, and this is only possible if we have reasonably reliable prognoses of the economic trend.
Another criticism against the Keynesian proposals was furthermore that national debt leads to undesirable secondary effects on monetary stability and growth. The increases in demand triggered by the deficit of the national budget will in any case partly fizzle out in price increases, since the economic development in the individual branches of the economy is asynchronous and since therefore price increases already occur in individual areas long before mass unemployment could have been eliminated in the rest of the economy.
The further fact that a Keynesian employment policy makes it necessary for government measures to follow the course of economic activity in a go-stop cycle leads to additional uncertainty about the interest rate development, which in turn causes the risk that long-term investments decline, and thus long-term growth is impeded.
In the post-war period, efforts were made to partially reduce these efficiency deficits in financial policy instruments. It was provided particularly that tax increases could also be implemented by decree for economic policy reasons and would only have to be approved by parliament subsequently.
02nd Is a reflation sufficient?
During the global economic crisis, the question was discussed as to whether it was sufficient for the state to stimulate the economy once in the sense of pump priming and thus lead out of the economic depression, or whether a permanent budget deficit was necessary in order to ensure that the aim of full employment was not immediately questioned again when the state discontinued its boosts to the economy.
The multiplier theory developed by Keynes speaks actually for the fact that permanent intervention is needed to achieve lasting success by way of fiscal policy. Only if there is a permanent increase in government spending there will be a sustained increase in the domestic product. For each period applies the fact, that full employment can only be expected if and as long as the overall economic demand corresponds to the possible supply of goods at full employment.
In order to ensure that it is sufficient to have a sustained positive impact on the economy by means of one-time economic stimulus measures, it is necessary that an equilibrium path exists at all and that the only a deficiency is that the market lacking the ability to eliminate the existing imbalance (excess supply) by itself. The following graph illustrates this connection:
The intersection of the red supply curve and the blue demand curve shows at which domestic product (Xgl) full employment is reached. The course of the demand curve shows that an equilibrium point exists and that only the market has no equilibrium tendency, thus is incapable of heading for full employment on its own from the current position (X0). In such a situation, it would be sufficient for the state to be able to raise demand to the level of full employment with a one-time initial assistance, and that this full employment income would also be maintained in the further periods even if the state were to stop providing further economic stimuli.
The only critical point is that Keynes assumed a different situation. His assumptions correspond to the following diagram:
Here, it is assumed that the goods market equilibrium is situated at the current domestic product with underemployment. Now, if the state were to raise the domestic product to the level at which full employment prevails (Xgl) by means of one-time economic stimulus measures, there would be an excess supply on the goods market and due to this excess supply the market, left to its own devices, would again fall back to the initial level (X0). Permanent full employment could only be achieved if the state would increase overall demand on a permanent basis by means of a corresponding budget deficit. Pump priming would not be sufficient to guarantee full employment permanently.
But within the framework of the further development of Keynesian theories there are also theses that speak for the fact that one-time economic stimulus measures are sufficient and that the market can maintain the level of full employment by itself when the level of full employment is reached with the help of the state.
One-time economic stimulus measures would improve the investment climate in the short term and would therefore also have long-term effects, since the investment volume would then be increased in the long term. Primarily the consideration of the acceleration principle speaks for the success of even one-time interventions by the state. Paul Samuelson has shown that the interaction of multiplier and accelerator can lead to longer lasting economic upswings.
Here, the black line indicates the development in deficit-financed public expenditures. In the graph, the budget deficit is assumed to be reduced to zero already in period 2. The red line marks the induced increase in income shown in the multiplier theory. The graph shows clearly that although the multiplier effect also persists even if public expenditure growth is stopped, but that this influence approaches zero very soon.
The orange line shows how investment responds to income growths, at first the investment increases strongly, but then declines again as income has peaked. Finally, the blue and pink curves show how income and consumption develop in response to both the multiplier as well as the accelerator.
The graph makes clear that a one-time economic stimulus triggers a fairly long-lasting upswing, so that this economic upswing is not in danger even if the state reduces its financial support again after the first period. However, this graph also shows that it is hardly possible to avert the danger of cyclical unemployment with a one-time boost for all time. Beginning around period 7, income and employment will again fall strongly, so that in the long term again state stimulus measures will be necessary anew.
For this reason, the fundamental criticism of a Keynesian full employment policy remains. It does not get down to the actual causes of mass unemployment and therefore limits itself to treating the symptom.
03rd Built-in flexibility versus autonomous economic policy
We now turn to another contentious point within Keynesian employment theory. Should the state autonomously adjust the level of the budget deficit to the required level in each period or should be tried to develop certain mechanisms, so-called built-in stabilisers, which automatically ensure that the deficit of the national budget reaches the required level without permanent intervention by the government?
Initially, it is necessary to define the concept of built-in stabilisers: a built-in stabiliser triggers the desired influence on the economy without further autonomous decisions by the state. The tax progression in the income tax rate would serve as an example.
If we assume an economic boom, it is necessary to dampen demand in order to avoid inflation: As income rises in this phase of the economy, the average tax rate rises, too. And as the tax rate rises, the consumption ratio in relation to gross income decreases finally. Thus, a stabilising effect occurs automatically - without direct intervention by the state.
Mentioned as a second example of a built-in stabiliser is the public unemployment insurance system. If we assume a recession, demand stimulation is necessary to avoid unemployment: since income decreases, the contribution sum also decreases; as unemployment rises at the same time, the amount of expenditure rises also, and as the amount of expenditure rises, finally the demand rises. This means that here, too, the economy will recover automatically.
Now, what are the advantages of a built-in stabiliser? The effect occurs immediately (there is no inside lag). We have seen, though, that it is precisely the fact that a lot of time elapses from the moment that, for example, mass unemployment occurs until the moment that the policy measure enters into force that makes these measures ineffective.
Built-in stabilisers also have disadvantages, however. These disadvantages consist in the fact that no consideration of the current situation is possible, here. But we must assume that every economic situation is unique to some extent. Walter Eucken said, for example, that precisely because of this uniqueness there could be no general economic theory.
This uniqueness results from the fact that each economy consists of a multitude of markets, with each market having a multitude of features which differ from one another in the individual case. But if each individual economic cycle has different features, then it is likewise not possible to consider all eventualities in a legislative framework.
And in this case, an autonomous decision by the state could very well consider many more differences in the structure of the markets and thus achieve much more appropriate solutions than if an automatism applied to many economic phases were to relieve politicians of this decision.
Now, let us take a more detailed look at the effect of a public unemployment insurance as a built-in stabiliser. The starting point would be an I-S model, with the intersection of both curves (the private investment function and savings function) at an income below full employment.
The introduction of a public unemployment insurance now leads on the one hand to a shift in the savings curve upwards parallel to the increase in the premium income (b), and it now becomes the curve of the discontinuation of purchasing power (S+b), and on the other hand leads to a shift in the investment curve due to the granting of unemployment benefits (AG); the curve of the absorption of excess buying power (I+AG) takes the place of the investment curve now.
In the case of full employment, the curve of the absorption of excess buying power would coincide with the investment curve, since it is assumed that no unemployment benefit is to be paid. However, the lower the income and employment turns out, the larger is the amount of unemployment benefits and the more deviates the curve of absorption of purchasing power upwards from the investment function.
Even if the unemployment benefit were 100% of the previous wage income, the unemployment benefit curve would only rise by 45°. But since unemployment benefit must be significantly lower than the previous wage income in order to avoid abuse, the inclination of the unemployment curve is much lower than 45° in reality.
If the unemployment benefit were equal to the previous wage, the unemployed would not have enough incentive to take up a new job. It would be more advantageous for them to receive unemployment benefit at the level of their previous income without having to make the effort of a gainful employment.
I: investment expenditure
S: savings amount
Y0: national income without unemployment insurance
YArb : national income by granting unemployment benefits
Yvoll : national income with full employment
b: premium sum for unemployment insurance
AG: sum of unemployment benefits
The two lines, which are starting from the point of full employment and the investment sum and which are pointing to the top left, show the course of the unemployment benefit in relation to the national income, whereby the 45% line would apply for the case that the unemployment benefit corresponds to 100% of the previous earnings, while the line < 45° shows the course of the unemployment benefits in the case that the unemployment benefit, as in reality, is clearly below the previous incomes of the employees.
Let us assume the market equilibrium before the introduction of unemployment insurance. It is at Y0, at the intersection of the investment curve and the savings curve. The introduction of unemployment insurance causes now the new market equilibrium to be at the intersection of the new curve of purchasing power discontinuation with the purchasing power absorption curve, namely at YArb. The national income and with it the employment has thus increased.
But this graph shows also that full employment can never be achieved in this way alone. The new intersection is clearly at a national income that does not yet guarantee full employment, and this would even apply in the case that the unemployment benefit was to correspond to the previous wage income of the now unemployed.
04th Anticyclical versus compensatory financial policy
A further point of contention within the Keynesian employment theory was the question of whether fiscal policy was to be oriented anticyclical or compensatory.
We begin with the definition of anti-cyclical policy: in an economic downturn, budget deficits are realised, whereas in an economic upswing budget surpluses are realised, whereby the size of the (positive or negative) balance must correspond to the size of the market imbalances caused by the economic cycle.
In this context, the assumption is made that in the long term, i.e. over an entire economic cycle, supply and demand correspond and are only asynchronous in the short term. During an economic downturn demand lags behind supply, while during an economic upswing demand exceeds supply.
But for this reason, the budget can also remain balanced in the long term, which means beyond an economic cycle, only in times of a downturn would the national budget have to show a deficit, whereas in times of an upswing it would have to show an equally large surplus. Unemployment is therefore seen here as an exclusively cyclical and not secular problem. Regarding to the demand for an anti-cyclical policy, however, it would be critical to point out that there are no regular economic cycles.
In contrast, a compensatory policy provides that in each period the state replaces a lack of private demand by equally high budget deficits but compensates for excessive private demand likewise by equally high budget surpluses.
In contrast to the anti-cyclical policy, there are doubts as to whether the economic swings correspond to each other and that therefore a compensation of supply and demand occurred over an entire economic cycle. Rather, it was to be expected that there would be a lack of demand for secular reasons also in the long run, so that a deficit in the state budget would also be necessary in the long term.
But precisely because the fight against unemployment was not a problem of the economic activity at all, it was of importance which economic phase we are currently in for determining the required level of the deficit in the national budget. Even in times of an economic upswing, it must be reckoned with the possibility that the investment volume of the enterprises will not be sufficient to compensate for the loss of consumer demand which is triggered by an increase in savings.
Only in a rapidly growing population would the willingness to invest voluntarily be sufficient to match the savings amount. The two most important motives for the investment were that workplaces and housing would have to be built newly in the event of population growth. If, on the other hand, the population stagnated, these two motives would drop, and the investment amount would no longer correspond to the savings amount. This is particularly true because the savings rate rises with increasing per capita income. Since, in connection with the stagnation in population development, the investment volume is declining more and more and, the savings amount is declining simultaneously, the unemployment caused by insufficient demand is a secular, thus persistent problem.
The demand for a balance of the national budget thus requires that in each period the total demand corresponds to the supply at full employment. In a national budget, it should be determined for each period to what extent the difference between investments and exports exceeds the sum of savings and imports, and the balance of the state budget should be adjusted in such a way that the total effective demand in each period corresponds to the total decommissioning (S, Im, T).
AST: public spending
T: tax amount
I: investment amount
S: savings amount
Ex: export revenue
Im: import expenditures