1.3 Thinking like a state about economic contradictions: Tinbergen

A draft thesis section. This is a draft of an unfinished document, please don’t quote without getting in touch first. Quoting in blogs is fine.

Previous section / Next section

Jan Tinbergen connects to the technocracy, looking rather Dutch by the way.

Jan Tinbergen: "Get me the technocracy! Class conflict has arrived!"

The fact that the economic ‘problems’ appearing for policy resolution are plural raises the prospect that the demands made on the state pull in different directions. Even if the various branches of economic policy are pulled together as if to form a unified strategic actor, there is no guarantee of an optimal strategy capable of resolving the contradictions underlying the symptomatic problems. Rather, the contradictions reappear in the political realm. The original Keynesian objective of ‘full employment’ was in reality only one among several objectives governments attempted to steer the economic system towards, including rapid growth, ‘external balance’, and, of course, price stability. A number of objectives, then, but only one economic system, and within the system the objectives were not independent of one another. Some were very much in tension with others.

In this thesis I am concerned with the tension between price stability and other goals: first why it appeared, given the structures of the economy and state, and second, why it – originally a subsidiary goal – increasingly came to dominate policy strategy.

Australian policymakers became increasingly aware over the course of the post-war period of the tensions between their objectives, as I discuss in Chapters 3 and 4. Such a sense of contradiction and trade-off pervades the 1965 report of the Committee of Economic Enquiry, commissioned by Prime Minister Robert Menzies precisely to investigate the tangled mess policy seemed to be in following the credit crunch and recession of the early 1960s:

It must be recognised that the attainment or near attainment of any one of the seven objectives of economic policy may make the attainment of others more difficult… Thus, the nearer an economy is to full employment, the more difficult it is to achieve stability of costs and prices. On the other hand, stability of costs and prices makes for external viability, although action to achieve external viability, for example, by exchange devaluation, may operate against stable prices. In short, all objectives cannot be ‘maximised’ simultaneously… [Vernon et al, 1965: 46]

Similar conclusions had been reached in other countries, and the 1950s had seen the emergence of a new genre of economic theory examining precisely this issue. Perhaps its most clear and influential expression came from Dutch policymaker and academic economist Jan Tinbergen [1955, 1966]. In Tinbergen’s work we get something approaching a picture of how the economic policy ‘agent’ might think if it really did have a brain. The crisp formal models are still far from the messy uncertainty of diagnosis and action in the real-world economic bureaucracy. Nevertheless, his models capture with some clarity the structure of the economic policy problematic, opening the way for analysis of its contradictions and structural development – even if this aspect was not often followed up in the genre he established, with its ‘policy mix’ and ‘policy reaction functions’.

Tinbergen’s [1966] project is essentially about working out the interconnections between multiple economic objectives and assessing how, in consequence, policy instruments must be co-ordinated. This reflects the pseudo-agent-making process de Brunhoff [1978: 83] refers to:

It is not possible to define an ‘economic policy’ simply by enumerating and adding together these various elements [– isolated institutions and goals]. A certain necessary homogeneity, affected by the reduction of these elements to monetary flows and interdependent sectors, is also required. But this alone is not enough. One cannot begin with a coherent project resulting from a global objective, towards which the various complementary policy measures would be combined. Such a model presupposes the existence of the state as subject, instead of showing how, in certain circumstances, the state ‘can be called upon to function as a subject’ of economic policy, even is the policy in question does not in any sense have a coherently defined totality of objectives or the mechanisms able to attain them.

Compare Tinbergen’s complaint of – and remedy for – a “tendency to an incoherent treatment” in economic policymaking:

Measures regarding various instruments are taken separately, often at different moments and without much co-ordination. This tendency is to some extent based on the belief that there is a one-to-one correspondence between targets and instruments, that is, that each instrument has to serve one special target. Taxes and government expenditure are thought to be relevant to financial equilibrium, wage rates to employment, exchange rates to the balance of payments and so on. The interdependence is neglected or underestimated… It would seem that a better approach has now become possible. It is no longer necessary to neglect the interrelations, and a simultaneous consideration of all targets and instruments, as well as their quantitative relations, should be considered. [Tinbergen, 1966: 49]

His procedure is to build a formal model of the economic system which incorporates all the goals authorities pursue. He starts from a simple model in which only employment is targeted, and gradually adds in extra goals to show the additional complexity at each step. In each model, the relationships are corralled into a list of

  1. target variables;
  2. instrument variables directly controlled by policymakers;
  3. exogenous data neither controlled by the policymaker nor targets in themselves; and
  4. a system of equations linking the variables together.

Targets are either absolute values (e.g., absolute price stability, or a given rate of inflation, or a rate of unemployment), or ‘flexible’ values to be maximised or minimised given other conditions. Targets might be ranked, as they are in practice, so that one overriding goal has an absolute value while another is pushed as far as possible given that limitation.

Instrument variables are assumed to be set by authorities. Depending on which variable is identified, this might stretch reality quite far: for example, where Tinbergen treats aggregate expenditure as an instrument, he grants far more control to fiscal and monetary policy than they actually have. It would be more realistic to identify only those components or determinants of aggregate expenditure which are genuinely under state control as instruments: the fiscal stance and the base interest rate, for example.1 Other influences on aggregate demand could then appear as exogenous data. Where to draw the line between (2) and (3) depends on the real conditions in which policy works.

In fact much of the history of policy development involves not the working of a fixed system of instruments, but of attempts by policymakers to extend their effective reach, and counter-developments somewhere or other in the socio-economic sphere that frustrate them. All elements of ‘exogenous data’ are not equal: some are truly beyond potential state influence, while others are just beyond reach, more predictable or more susceptible to being coaxed in one way or another, holding out the prospect of pulling them fully into the policy orbit. Rather than a tight distinction between ‘instruments’ and ‘exogenous data’ then, it is better to think of a chain of causal influence between the instrument directly set by authorities and the ultimate target, with some elements closer to direct control than others. Tinbergen does not put it exactly in these terms, but he does make a distinction between ‘quantitative policy’, involving the working of an established system of policy transmission, and ‘qualitative policy’, which aims to reshape the system itself.2

Section to be continued.

1Even these variables grant too much control to the authorities – in reality fiscal stance is somewhat unpredictable, given its two-way causal interaction with private flows of expenditure, and even the stability of the central bank’s control over the base rate can come into question, while its effectiveness over other rates and quantities is uncertain and shifting.

2In fact he distinguishes between ‘qualitative policy’ “in which the structure of the economy is changed” and ‘reforms’ in which changes affect “spiritual aspects or relations between individuals”, but it seems to me that this is a difference of magnitude rather than kind. Examples he gives of the former include rationing of goods or foreign exchange, welfare measures, changes to tariff structures and anti-monopoly legislation, and his examples of the latter include the introduction of a full-scale social security system, nationalisation and industrial democracy. [Tinbergen, 1966: 149]


Suzanne de Brunhoff [1978]: The State, Capital, and Economic Policy, translated by M. Sonenscher, Pluto Press, London.

Jan Tinbergen [1955]: On the Theory of Economic Policy, 2 ed., North-Holland, Amsterdam.

Jan Tinbergen [1966]: Economic Policy: principles and design, North-Holland, Amsterdam.

J. Vernon, J. G. Crawford, P. H. Karmel, D. G. Molesworth and K. B. Myer [1965]: Report of the Committee of Economic Enquiry, Commonwealth of Australia, Canberra.


The URI to TrackBack this entry is: https://scandalum.wordpress.com/2009/06/04/1-3-thinking-like-a-state-about-economic-contradictions-tinbergen/trackback/

RSS feed for comments on this post.

One CommentLeave a comment

  1. […] Previous section / Next section […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: