Are Keynesianism and Neoclassical economics antipodes?
by Heiner Flassbeck
Keynesianism and neoclassical theory are often seen as the two main approaches in economic theory. Many assume that the large gap that exists between the two approaches or schools can be bridged by a middle oft he road approach based on pragmatic assumptions and empirical research. This assumption is incorrect.
In meetings with students who are critical of the economics that are being taught today by mainstream theory and who are interested in Plural Economics, Real World Economics and other heterodox attempts, one often meets the type of the doubter. She or he understands that the ruling neoclassical doctrine fails to explain economic phenomena. These students, therefore, look for better explanations. However, Keynesianism, which would be the obvious candidate for an alternative theory, is also being seen with some skepticism. To these students Keynesian economics is the antipode to neoclassical economics and they try to find the middle of the road, equidistant from one and from the other.
This antipodal view is not justified because the two “extremes” do not mark the two ends of a straight line. In the antipodal view neoclassical economists consider the market as the main organising principle of society. The role of policy-making lies in engineering conditions in such a way that market forces can fulfill their work in an efficient way. For Keynesians, on the other hand, the concept of market failure stands at the centre of the analysis. The state has to play an essential role in economic policy. What is more likely than that a young and critical student decides to keep a healthy distance from both approaches? Presumably, she or he assumes, the truth lies somewhere in the middle.
Furthermore, positions that seemingly fit perfectly into this dichotomist view of the world can be found everywhere. To the neoclassists, all prices must react flexibly to changes in supply and demand for the market economy to function. Keynesians, on the other hand, tend to build models with rigid wages and prices and defend the rigidity of these wages.
Saving and Investment
Neoclassical economists consider the transformation of savings into investments as the most important of all transformation mechanisms in a market economy. Savings play an essential role, as free and (rational) decision-making consumers decide whether they prefer to spend their income today or save it in order to spend it at a later date. Keynesians argue that the decisions of private households are indeed crucial for investment, but that their savings are primarily the result of investment of company and government. Expressed as a dichotomy: savings determine investment or investment determines savings.
With regards to the external counterpart to domestic savings, which is saving of the national economy or the surplus of the current account, the positions also appear to develop in exactly the opposite direction. Neoclassists emphasize the importance of capital flows for the creation of such balances with the main determinants being the autonomous decision of consumers to either save or spend. This is based on the famous criterion of micro-foundation and was turned into the main criterion of scientific work in the field of economics. However, consistent Keynesians see things exactly the other way round: capital flows adapt to the flow of goods, which are, in turn, determined, among other things, by the competitiveness of a country (in essence the undervaluation or overvaluation of its currency) (I come back to this in Part 3).
There is no centre
Because the antipodes appear to be so obvious, critical students are often looking for a compromise between the two. This is understandable, but it is a mistake. There is no middle ground between Keynesianism and neoclassical economics. Nor is there a critical third body of work outside the two main schools that can aspire to increase our understanding of the functioning of a market economy. It is, of course, possible to move towards the margins of scientific endeavour and deal with issues that both theoretical systems have neglected, but this does not invalidate the point that whoever wants to participate in the decisive debates on global and national economics must choose sides. Otherwise it will become difficult to say anything relevant.
The above-mentioned statements sound like theoretical antipodes. In reality, they are the result of systems that are based on very different assumptions. From my point of view, the willingness to integrate macroeconomic indisputable relationships, epistemological reasoning and solid empirical results into the theoretical system are the decisive points, which make the two “theories” completely methodologically incomparable.
Let us have a look at the most politically important example, the current balance of some large economies and their causes. Paul Steinhardt pointed out a few days ago (see here) that a clear understanding of the relevant contexts is needed to arrive at reasonable conclusions. The so-called financial balances mechanics (that is, a logically indisputable macroeconomic relationship) is indispensable, but it is not sufficient to answer the really decisive question, which is one of cause and effect.
Capital flows and trade balances
Steinhardt cites Jeffrey Sachs and Martin Wolf who both support Hans-Werner Sinn’s thesis that current account surpluses depend on the “savings” of a national economy, i.e. high propensities to save induce current account surpluses and vice versa. However, by definition it is true that economies that have current account surpluses spend less than they earn, while economies with current account deficits spend more than they earn. In a contribution for Social Europe that deals with the German surpluses, another important economist, Barry Eichengreen, writes that:
“Back in the real world, the explanation for Germany’s external surplus is not that it manipulates its currency or discriminates against imports, but that it saves more than it invests. The correspondence of savings minus investment with exports minus imports is not an economic theory; it’s an accounting identity. Germans collectively spend less than they produce, and the difference necessarily shows up as net exports” (see here).
To put it plainly, this is just frightening. Eichengreen is a world-wide known, highly traded economic historian, who explicitly sees an identity as a statement of content and obviously does absolutely not understand that the decisive and explanatory factor behind these identities are the movements of income. The whole core of the Keynesian revolution in economic theory has passed the man!
At this point, those observers with a critical mind have to ask an essential question about this indisputable accounting relationships: if it is by definition correct that economies which generate surpluses spend less than they earn and if this behaviour is called “saving,” then this means that those who save have surpluses and those who do not save have deficits! This statement, however, is nothing else than the repetition of the accounting logic, this time by using the word “saving.” Consequently, the statement (and any empirical research that is based on it) is completely empty. On the same level, “as a Keynesian,” one could only say that economies that export more than they import have a current account surplus and vice versa.
While the latter statement is being dismissed as a triviality, the first, neoclassical, statement is being used like a theory and is considered a statement with substance, although it means of course precisely the same as the second one and has no substantive content whatsoever. Obviously, the problem is that the word “saving” creates the fiction of relating to the behaviour of private and public households. But accounting cannot explain the actions of these actors. The truth is that this so-called „saving“ has nothing to do with the concrete behaviour, preferences or actions of the private or public actors.
Not everyone can be a saver
The previous statement is easy to verify. It is sufficient to ask how the relations between savings and investments can be reconciled with micro-economic logic at the global scale. We know quite well from the accounting logic (and – not to forget – from healthy common sense) that there is no way for the world as a whole to save in the sense mentioned above, which means to spend less than the earnings and to create in this way a current account surplus. The net saving as well as the net debt of the world are always exactly zero. This simple fact has a devastating consequence for the attempt to find a micro-economic foundation as is requested by the neoclassical approach.
All countries of the world can have the desire to be net savers in order to provide in this way for their future (just as the surplus is portrayed in all seriousness in Germany as they key to a prosperous future), the problem is that for logical reasons this is absolutely impossible. All the squirrels of the world can save (put aside) nuts, but it is impossible for all countries to have a surplus. This means that the attempt to find a micro-foundation is absurd. What we need for a correct analysis is a macro-foundation, the recognition that there are necessary conflicting desires of countries. With this the dichotomy itself also collapses. Neoclassical economics, dominant as it may be, is not an alternative to Keynesian economics: it is not a consistent approach and has to be rejected a priori.
Imagine that all countries in the world follow similar savings policies. For example, they all wish to save more in order to provide for the future. If this is the case, there have to be one or more mechanisms that ensure that such (inconsistent) saving policies are brought in accordance with the logical necessity for the world to achieve a perfectly balanced current account. And this fallacy of composition is just ignored by most economists. The struggle for current account surpluses and positive net savings is a zero-sum game for all countries in the world (unlike total trade as such). This means that some countries necessarily have to abandon their ambition to accumulate current account surpluses. But how can this be done if everybody wants to save?
Zero-sum logic
Neoclassical economics has no answer to this question. It ignores this blatantly obvious problem. Therefore, it does not compete in the race of serious economic theories, because the capital flows that originate from current account surpluses do not have a zero-sum character. However, many Keynesians very often also fail to grasp this absolutely essential point. Instead of providing a logically compelling explanation, they consider many different factors that can lead to current account imbalances, such as changes in terms of trade, differences in the growth rates of the countries and differences in competitiveness between countries.
However, even if the growth rates of all countries were the same, there would still a mechanism be needed that brings the sum total of the current account balances of all countries in the world to zero. It is plain logic again that such a mechanism must have a similar zero-sum character as the balances itself: every gain has to correspond to a loss as every surplus corresponds to a deficit. The two most important mechanisms of this kind are changes in the terms of trade, that is, changes in import prices compared to export prices, and – it is not exactly the same – changes in export prices (expressed in international currency), that is, real exchange rate changes, which are the most important expression of the competitiveness of nations.
Price ratios are crucial
This obviously means that in a consistent global analysis, only price ratios in the widest sense can be responsible for current account balances. The changes in these price ratios either reflect real changes (for example, increasing scarcity of raw materials) or, as real exchange rates changes, describe a process that explicitly regulates conflicts between similar ambitions or desires of countries. How this happens and whether it is rational or chaotic remains a completely open question.
The normative answer that economic theory can offer for the determination of real exchange rates is straightforward: since there can be no “ambitions” or claims in international relations that one country can legitimately enforce against the will of others, the only rule that independent and equal states can agree upon is that real exchange rates and the conditions that regulate competition between nations should not change at all. At the same time this is a miraculous micro-foundation because only under these conditions does competition between companies take place on the global level as it does within national states.
Constant real exchange rates
Consequently – and this is crucial – under such conditions, (nominal) changes in exchange rates must precisely compensate for inflation differences between countries. If this happens, at the international level everything is done that can be expected to make the process of balancing current account balances as conflict-free as possible. This was and is the perfectly correct and indisputable core of the idea behind the global currency system of Bretton Woods. It was this core, which intellectually clearly belonged to Mr. Keynes, that proved to be a decisive advance above and beyond the concrete design of the former dysfunctional economic system which un-ruled the world in the 19th and the early 20th Century.
The acceptance of this rule has enormous consequences for the coexistence of the nations. First and foremost, it prevents any attempt of one nation to compete as nation against others. Each country is, so to speak, thrown back onto its own economic endowment. A country can strive to increase its productivity, but it has to use any increase to grow domestic incomes because a strategy of undercutting other countries is impossible. To lower national inflation by putting political pressure on nominal incomes would never lead to improved competitiveness because of the unavoidable appreciation of the currency.
If countries work within monetary unions or in a system of fixed exchange rates, no country is allowed to undervalue its implicit currency by letting wages grow below its productivity gains. Wages must rise at all times with national productivity and the rate of inflation. Without this minimum standard of rational behaviour of individual member states, a monetary union or systems of fixed exchange rates can never work. Germany, from the outset, has broken this rule within the framework of the EMU and thereby violated the core of the reason behind such a system.
These considerations prove that monetary cooperation between nations that want to trade with one another is unavoidable. The simplest form of cooperation would be to leave the price of currencies to a foreign exchange market, which always finds the right exchange rate and therefore ensures the existence of constant real exchange rates. But such a foreign exchange market does not exist (see, for example, here for an explanation). Therefore, institutional regulation that determines exchange rates, i.e. intergovernmental cooperation, is the only way to meet this minimum requirement for meaningful and efficient international trade. Neoclassical economists normally deny the necessity of such cooperation. But that is just due to their ideological bias towards a government free market ruled world.
The rules concerning the level of fixity in any exchange rate system can be determined in a rational way, according to the degree the member states in such a system are willing to recognise the principle of constant real exchange rates even without nominal exchange rates and how far they are ready and able to coordinate national wage policies. If the propensity to do so is not very pronounced, one has to ensure that monetary relations adapt as early and smoothly as possible to the resulting inflation differentials (see a proposal by Peter Bofinger and myself here – see Chapter IV).
S = I
Neoclassical economists do not generally reject the use of accounting balances. Surprisingly, one of the identities is all too frequently used and, above all, too easily. The equality of savings and investment, I = S is the neoclassical battle cry that shall prove that market forces always do right and that you can always rely on it.
That savings equal investment is always correct in the sense that the parts of income that have not been consumed have been invested. This is true because only these two categories of spending exist in the world as a whole. But put in this way, the essential of the relationship remains unmentioned and unexplored. In which way are savings and investments brought together in the course of time and, most importantly, how do changes in income interfere in the unfolding of this process? About this, the definition says absolutely nothing.
To find a solution to this problem a neoclassical author will behave quite differently from what a normal social scientist would do. A social scientist, who is not a priori interested in market processes, will wonder in which way and in which sequences of events, the income of an economy is being created and he will try to understand the roles of savings and investment in this process. Although he knows that, ex post, savings and investments must be equal that does not prevent him from thinking about the different processes that can generate income.
But this is exactly not the way a neoclassical economist approaches such a problem. He or she looks for a market process that matches the willingness of private households to save with the need of companies to invest. He would a priori try to find a price that can guarantee the necessary adjustments of corporate investment plans even when private households save either less or more for their individual provisions. If he finds such a market process and its price, the problem is considered to be solved because the market and the flexibility of the price will assure that S and I are always the same. The key price in the market for capital has been identified as the interest rate, which is supposed to decrease when households save more and to stimulate the investment of companies at the same time.
Whether the income of the national economy, which is the reason for all economic activity, rises or falls in the process driven by the interest rate is not crucial to the neoclassical economist because the development of income (the growth path) is considered to be given (from technical progress and other factors) and the adjustments coming from the capital market.
Is there a „natural“ order?
The role of the state for meaningful economic development is non-existent for a neoclassical economist. A biologist, who is looking for the symbiotic relationships between ants and leaf-frogs in the deepest jungle, would also not have the idea of introducing the role of humans in this process. The “natural order” in neoclassical economics is entirely possible without the state. Therefore, and again by a priori reasoning, there can be no solution to a problem which requires intervention by the state.
Once the natural order has been “found” or invented, in our case the capital market, there is no need to further explore the laws that govern the system. Only restrictions for the freedom of the capital market through state intervention limiting price flexibility and preventing the market from ensuring the efficient transformation of savings into investments, can prevent on optimal result. Whether this capital market interest rate mechanism exists at all under realistic conditions or whether it can exist in a paper money economy without the state and the central bank are questions that neoclassical economics never ask. They do not have to ask such questions because they have done what was their intention: to find a “solution” that looks like being a pure market solution.
However, a field rejecting positive analysis of real world relationships declaring such relationships a priori to be irrelevant is not a science in the sense that we would normally define science, based on classical natural sciences. Neoclassical ideas have become unscientific exactly with the attempt to adopt scientific methodology by the mathematical formalization of economic relationships in a general equilibrium framework. It is the general equilibrium framework that demands a market and a flexible price for each and every problem at the micro and at the macro level and that is why it has to be flatly rejected. It is a pure glass bead play (see here here for a contribution from mine from the year 2004 on this topic) because almost nothing of what was postulated with this attempt as market economy finds a positive empirical confirmation in our societies.
Even worse: as soon as deviations from the model are shown in reality, neoclassicism tends to become immediately normative by calling on society to approach the market model instead of adapting its own findings.
Over time, it has become abundantly clear that Keynesians made a major strategic mistake to consider neoclassical economics as a scientific counterpart that has to be opposed. Instead of characterizing neoclassical economics had from the very outset as a normative structure, which serves no scientific purpose, they took it on as a science. But no science can ever win the confrontation with an ideological superstructure. The neoclassical attempt is not directed to make genuine scientific progress, but to defend its own position at all costs and even if this cost comes with the high price of inconsistency no one cares. The “general theory,” which Keynes attempted to write, was not ‚general‘ because neoclassicism was not a special brand of economic theory, but a normative structure.