Economic Growth Second Edition
Figure 1.11 The Solow–Swan model with technological progress
Download 0.79 Mb. Pdf ko'rish
|
BarroSalaIMartin2004Chap1-2
- Bu sahifa navigatsiya:
- 1.2.13 A Quantitative Measure of the Speed of Convergence
Figure 1.11
The Solow–Swan model with technological progress. The growth rate of capital per effective worker (ˆk ≡ K /LT ) is given by the vertical distance between the s · f (ˆk)/ˆk curve and the effective depreciation line, x +n +δ. The economy is at a steady state when ˆk is constant. Since T grows at the constant rate x, the steady-state growth rate of capital per person, k, also equals x. 1.2.13 A Quantitative Measure of the Speed of Convergence It is important to know the speed of the transitional dynamics. If convergence is rapid, we can focus on steady-state behavior, because most economies would typically be close to their steady states. Conversely, if convergence is slow, economies would typically be far from their steady states, and, hence, their growth experiences would be dominated by the transitional dynamics. We now provide a quantitative assessment of how fast the economy approaches its steady state for the case of a Cobb–Douglas production function, shown in equation (1.11). (We generalize later to a broader class of production functions.) We can use equation (1.39), with L replaced by ˆL, to determine the growth rate of ˆk in the Cobb–Douglas case as ˙ˆk/ˆk = sA ·(ˆk) −(1−α) − (x + n + δ) (1.41) The speed of convergence, β, is measured by how much the growth rate declines as the capital stock increases in a proportional sense, that is, β ≡ − ∂(˙ˆk/ˆk) ∂ log ˆk (1.42) Growth Models with Exogenous Saving Rates 57 Notice that we define β with a negative sign because the derivative is negative, so that β is positive. To compute β, we have to rewrite the growth rate in equation (1.41) as a function of log (ˆk): ˙ˆk/ˆk = sA ·e −(1−α) · log(ˆk) − (x + n + δ) (1.43) We can take the derivative of equation (1.43) with respect to log (ˆk) to get an expression for β: β = (1 − α) · s A · (ˆk) −(1−α) (1.44) Notice that the speed of convergence is not constant but, rather, declines monotonically as the capital stock increases toward its steady-state value. At the steady state, s A · (ˆk) −(1−α) = (x + n + δ) holds. Therefore, in the neighborhood of the steady state, the speed of conver- gence equals β ∗ = (1 − α) · (x + n + δ) (1.45) During the transition to the steady state, the convergence rate, β, exceeds β ∗ but declines over time. Another way to get the formula for β ∗ is to consider a log-linear approximation of equation (1.41) in the neighborhood of the steady state: ˙ˆk/ˆk ∼= −β ∗ · [log(ˆk/ˆk ∗ )] (1.46) where the coefficient β ∗ comes from a log-linearization of equation (1.41) around the steady state. The resulting coefficient can be shown to equal the right-hand side of equation (1.45). See the appendix at the end of this chapter (section 1.5) for the method of derivation of this log-linearization. Before we consider further the implications of equation (1.45), we will show that it applies also to the growth rate of ˆy. For a Cobb–Douglas production function, shown in equation (1.11), we have ˙ˆy/ˆy = α · (˙ˆk/ˆk) log Download 0.79 Mb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2025
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling