The Theory of Economic Policy and the Link between the Two Pillars
1.5.1 Developments after World War II and Extensive State Intervention
After World War II, a number of the actual features of the economic and social systems in advanced countries had changed or the changes looming before the war became evident with clarity (Thomas 1994: chap.
4). In particular, new social groups had emerged or had become more powerful, and growth of big business had brought with it growth of big labour, which was to acquire greater voice and power as conditions for full employment were progressively restored after reconstruction of factories and infrastructure.Government intervention was much more widespread and penetrating. To a large extent, it started from the Keynesian precepts aiming first at full employment. Since fears of stagnation were very diffuse, following the World War II experience, a new policy target emerged - growth. For the most open economies, another issue - the balance of payments - began raising concern before the Marshall Plan and when the relief of American aid ended. The international economic institutions created at Bretton Woods in 1944 had devised proper rules to avoid excessive capital movement, ensure symmetric adjustment by ‘creditor’ and ‘debtor’ countries, regulate free trade in a way so as not to impair the welfare state provisions and ensure international liquidity. In practice, domestic policy adjustments were necessary only for debtor countries. For those countries, equilibrium of the balance of payments and, with it, the need to ensure a proper evolution of internal prices - which had soared just after the war as a result of a lack of supply and again rose in the 1950s as a consequence of sustained growth - became part of the set of macroeconomic targets. The four targets - unemployment, inflation, growth and balance of payments - made up the ‘magic square’ later introduced by Kaldor (1971).
Keynes had insisted on the use of different policy instruments (both monetary and fiscal policy), and the war experience had shown the possibility of adding price and income controls to these, while the International Monetary Fund (IMF), even within a fixed exchange rate system, allowed for parity adjustment.
The need then arose for (1) managing the different policy tools in a way that took account of the multiple interrelations between macroeconomic targets and policy instruments, aided by input-output analysis, national accountancy and econometrics, and (2) preparing a programme (a plan), possibly covering more than one year, as the Marshall Plan itself required.
Some conceptual tools that had been devised in the interwar period were useful for empirical analysis and policy. These were the use of national accounting (including introduction of the concept of gross domestic product (GDP), improving on that of national dividend used by Pigou) as an instrument for practical implementation of Keynesian policies, the discipline of econometrics as a development of mathematical economics that permits empirically assessment of interrelations and the values of the system's parameters[7] (Tinbergen 1935). Construction of formal models to be tested against reality introduced the idea of the need for consistency of different public policies and the possibility to check their real effectiveness. This factor played an essential role in the development of the theory of economic policy. It then appears not so strange that the authors who first contributed to this theory (e.g. Tinbergen and Frisch) were among the founders of the Econometric Society and the journal Econometrica in the early 1930s. Finally, inputoutput analysis was introduced by W. Leontief (1941), who later adapted it to a very versatile policy analysis tool, especially useful for disaggregating macroeconomic models.
In addition to these developments, a contribution to the idea that governments could be successful in ruling economic systems - possibly even more than markets, which had been hit by the Great Depression - came from planned economies.
Such economies, in fact, had experienced high growth rates in the 1930s and following decades, which had an influence on the economic thought of many economists. Frisch, the first economist to devise the theory of economic policy, and his follower, L. Johansen, thought that national economic planning managed by well-trained economists was clearly superior to the market (Kffirgard, Sandelin and Sffither 2008), while being suspicious of the political architecture of Soviet society.10
Thus, after the war, the theoretical advances of the 1930s and the changes in the economic and social conditions, as well as in policy aspirations, prompted radical innovations. These changes occurred in the way that policymakers conceived their goals and scholars looked at economic policy, and in due time they resulted in the generation of a theory capable of dealing with these new issues, thus conceiving of economic policy as an autonomous discipline.
The first step in formulating economic policy as an autonomous discipline was the statement of the ‘theory of economic policy'. In the second step of this intellectual process, the theory of economic policy was associated with the existence of market failures. Each step will be detailed in the next two subsections.
1.5.2 Devising the Theory of Economic Policy
Frisch (1949, 1950, 1957) first formulated this theory.11 He had laid down some of its ingredients before World War II in participating in the debate on the economic feasibility of a planning scheme for socialist countries.[8] [9] His approach was in terms of ‘flexible targets'. The policymaker, in the same vein as a householder, should make a plan tending to maximise his or her preferences under the constraint of a set of equations describing the given behaviour of the economic system on the basis of technical, behavioural and - a later stage - institutional constraints. These constraints depend also on the value of the instruments that can be set by the policymaker. Thus, the government’s targets would be expressed in terms of ‘flexible targets’, as the values that could really be reached would depend on the constraints. The initial formulation of Frisch’s conception of the theory also was rather complicated because the constraints were expressed in terms of a system of input-output relations a la Leontief, with some devices that should allow overcoming non-substitutability between inputs. Tinbergen - who had built the first econometric models of the Dutch and US economies in 1936 and 1939, respectively (Hughes Hallett 1989) - drew inspiration from Frisch, as he recognised (see Tinbergen 1952: introduction) but followed a different approach. In fact, Tinbergen (1952, 1956) developed the theory in terms of ‘fixed targets’, suggesting that a number of instruments at least equal to that of targets should be available to the policymaker in order for him or her to be able to exactly reach the preferred set of target values (Tinbergen’s ‘golden rule of economic policy’). Frisch’s route was instead followed more closely by Theil (1956, 1964), who generalised Frisch’s formulation while making it simpler and more workable. The objective of the policymaker would be to minimise a loss function in terms of quadratic deviations from a set of target values for the variables of interest.[10] Minimisation would be under the constraint of the relations between targets and instruments expressing the behaviour of the economic system. 13 A plan of action for the policymaker can be drawn from this process of minimisation (or maximisation). In summary, a plan (or programme) is formed of three elements: targets, instruments and the model of the economy. Targets, goals or objectives have already been defined. The policymaker, perhaps with the help of an economist, as suggested by Frisch, constructs a map of ‘social’ indifference curves reflecting the preferences of society (a SWF defined over a set of ‘goods’ or a loss function defined over ‘bads’). This became the general way of setting a policy problem in a ‘parametric context'. In the case where the policymaker has a sufficient number of instruments - as indicated by Tinbergen's ‘golden rule' - the loss function is minimised, and the set of fixed target values is consistently reached by using appropriate values of instruments. In other words, under this condition of (at least) equality of the numbers of instruments and targets, the policymaker would control the economic system and direct it to reach his or her goals. Scandinavian and Dutch contributions to the theory of economic policy were not limited to Frisch, Tinbergen, Theil and other economists directly following their route. In fact, the Norwegian school of the theory of economic policy, founded by Frisch, had also other economists following his route (O. Aukrust, H. J. Kreyberg, B. Thalberg and, most importantly, Trygve Haavelmo and Leif Johansen) (Johansen 1977, 1978). Haavelmo was at the forefront of econometrics and macroeconomic models (see Anundsen et al. 2014) and had worked with the Cowles Commission. His works relevant to our topic are, unfortunately, mainly in Norwegian (such as Haavelmo 1966, but see Haavelmo 1965). To our knowledge, the most important contribution for tracing the roots and development of the Norwegian school is that of Bjerkholt (2005). 16 Lawrence Klein, who had worked with Cowles Commission from 1944 to 1947 with Haavelmo and many others under the direction of Jacob Marschack to build a model of the American economy, visited the Oslo Institute in 1947-48 to work with Frisch and Haavelmo (Visco 2014). He had published a very popular book on the Keynesian revolution, ‘had constructed the first macroeconomic model in [the] USA and demonstrated that the model, indeed, did better in forecasting than the experts in Washington' (Bjerkholt 2005: 15). Bent Hansen - a Danish-born economist who had graduated from Uppsala University - became an influential advisor to the Swedish government and a professor at the University of Stockholm (and the University of California at Berkeley) and offered an independent contribution to the goals/means theory of economic policy (see Hansen 1955; Erixon 2011). For a rather exhaustive review of the contributions to the theory of economic policy especially by Scandinavian and Dutch authors, refer also to Rey (1967), Hughes Hallett (1989), Jonung (1991), Kffirgard, Sandelin and Sffither (2008) and Acocella, Di Bartolomeo and Hughes Hallett (2013). 1.6