Professor Pasinetti, the Trend & the Cycle
Oleg Pavlov
Michael J . Radzicki
Department of Social Science & Policy Studies
Worcester Polytechnic Institute
Worcester, MA 01609
In 1962, Luigi Pasinetti published a model of the "Cambridge theory of income distribution" that
corrected a “logical slip" in Nicholas Kaldor's earlier formulation of the problem. Pasinetti's model
included two stocks of capital (capitalist-owned and worker-owned) instead of one, avoided
Kaldor's assumption that the workers' marginal propensity to save was zero, and demonstrated
that on a balanced growth path the rate of profit depends only on the growth rate of the labor
force and the capitalists' marginal propensity to save (the "Pasinetti theorem").
Earlier, in 1960, Pasinetti examined the multiplier-accelerator family of models and concluded that
they were unable to endogenously explain the interaction of the trend and the cycle due to their
aggregate character. More specifically, the macro-nature of the variables in these models
implicitly assumes that there is an economy-wide demand for a composite basket of goods that is
expanding in a proportional way over time. Pasinetti argued that such an assumption would be
true only in a theoretical economy in which economic growth was solely attributable to population
growth, with constant returns to scale, no changes in preferences, and no technological progress.
In such economies, entrepreneurs would be able to quickly learn the fixed preferences of the
population (and thus the fixed behavioral parameters of the system), produce to meet these
preferences, and steer the economy to a steady state path of equilibrium growth.
Pasinetti went on to argue that a more realistic model would allow for preference changes as per
capita income grows [from (say) subsistence goods to (say) durable goods to (say) luxury goods].
In such a system, entrepreneurs would make mistakes in their expectations about the future
composition of consumption, invest the wrong amount, and thus generate an interacting trend
and cycle. In a model of such a system, these mistakes imply that the behavioral parameters of
(especially) the investment function change over time.
The purpose of this paper is to extend Pasinetti's 1962 model by adding a behavioral
entrepreneurial expectations formulation to its structure. The extended model generates an
interacting trend and cycle and closely mimics U.S. macroeconomic data. Full Information
Maximum Likelihood estimation with Optimal Filtering (FIMLOF) is used to estimate the model's
behavioral parameters (including the parameters of the investment function), some of which
evolve over time. These results add value to the debate about the causes of the interaction
between the trend and the cycle and serve to blend ideas from Professor Pasinetti's two papers.
