Growth and Technological Leadership in US Industries: A Spatial Econometric Analysis at the State Level, 1963-1997

model expresses the growth rate of GDP per worker in each sector as a function of the
initial GPD per worker in the same sector. It reads as:
where Y_t^s is the initial GDP per worker, k the number of years in the sample period, s the
different sectors. Results of the unconditional convergence models are presented in Table 5.
In the unconditional convergence model, all the coefficients have the expected negative sign.
This indicates that there is β-convergence in all sectors. The rate of convergence varies
between 0.35% per year for the wholesale/retail trade sector and 4.02% for the
manufacturing industry. The combined sector “Total” has a convergence rate of 1.67%,
which is consistent with the prediction from cross-country and regional analyses (Abreu et
al. 2005). The observed rate of convergence in the manufacturing sector is also consistent
with the findings from Barro and Salai-Martin (1991).

[Table 5 about here]

In addition to the unconditional convergence estimation, we also test for the presence of
convergence clubs within each sector. Following the traditional approach, we divide the
sample into two groups based on the initial GDP per worker levels. We distinguish the
groups of high initial GDP per worker and low initial GDP per worker states on the basis of
above and below average GDP per worker in 1963. Following the “specific-to-general”
approach, we estimated the unconditional convergence model using General Moments (GM)

11

<<   8   9   10   11   12   13   14   >>

More intriguing information