Testing a Goodwin model with general capital accumulation rate

Abstract

We perform econometric tests on a modified Goodwin model where the capital accumulation rate is constant but not necessarily equal to one as in the original model (Goodwin, ). In addition to this modification, we find that addressing the methodological and reporting issues in Harvie () leads to remarkably better results, with near perfect agreement between the estimates of equilibrium employment rates and the corresponding empirical averages, as well as significantly improved estimates of equilibrium wage shares. Despite its simplicity and obvious limitations, the performance of the modified Goodwin model implied by our results show that it can be used as a starting point for more sophisticated models for endogenous growth cycles.

Testing Goodwin with a stochastic differential approach—The United States (1948–2019)

“…The recession dates reported are from the NBER.38 This comment echo results shown in Table3ofGrasselli and Maheshwari (2018). Namely, the over-simplistic investment function in the Goodwin model gives rise to dynamics for employment that does not match well the observed time series, while at the same time leads to an inferred equilibrium wage share that is at odds with the empirical average for that variable.…”

“…As is standard, the stock of capital, K, is assumed to accumulate with respect to investment, I, and to depreciate at a constant rate, δ, Although in Harvie (2000) the depreciation rate of capital was not included in the model, Grasselli and Maheshwari (2018), this parameter is assumed to be the mean value of the following time series By doing so, they found a value for δ G of 0.0521. 8 Using the above definition, δ depends on the level of the net capital stock, in particular on its initial value.…”

“…In previous attempts (Grasselli & Maheshwari, 2018;Harvie, 2000), the Goodwin model was estimated equation by equation using discrete-time techniques. 22 Each parameter was estimated separately using standard econometric tools such as an OLS, an error correction model or a vector error correction model.…”

“…This idea is discussed extensively inMohun and Veneziani (2006).29 This methodology is borrowed fromGrasselli and Maheshwari (2018).…”

The Broad Consequences of Narrow Banking

“…In all cases we use the base parameters shown in Table 2. Details on the parameters used for the wage, employment, and inflation parts of the model, namely α, β, η p , m, ν and δ can be found in Grasselli and Maheshwari (2018), whereas an in-depth discussion of the properties of the investment function and its parameters, namely κ i , i = 1, . .…”