In this, penultimate section it is appropriate to undertake a preliminary assessment of the empirical plausibility of our framework and its implications for “crowding out.” Our approach will be to combine the theoretical results summarized in Tables 1 and 3 with elasticity estimates for some key parameters, drawing upon the econometric findings of others for the latter.
The questions of greatest immediate interest pertain to the probable magnitudes implied for the other, unknown parameters of the model, and the consequent signs of the elasticity of nominal and real private R&D spending with respect to increases in public R&D spending. Recall that those magnitudes are in question only for the case we describe as “the long-run,” because in the “short run” the labor supply is taken to be fixed and the real volume of R&D resources invested (the number of R&D workers) has been shown to necessarily decrease when government expenditures rise. It should be evident that the computations reported here are rather conjectural in character; they are primarily illustrative in purpose, but offer a “reality check” on our model. The scientific sphere is the area into which the currect of capital investment shoulb be brought constantly. But when just a beginner you are not able to find enough money to start that’s why you may take speedy loans to conduct researches.
Our first avenue of exploration exploits the very interesting estimates that have been obtained by Austan Goolsbee (1998) for the short and medium to long-run elasticity of the R&D-worker wage with respect to R&D spending. Using a panel of scientists and engineers drawn from the U.S. Current Population Survey for the period 1968-1994, Goolsbee estimates that short-run wage elasticity for these workers with respect to nominal federal R&D spending is equal to 0.22, with a standard error of 0.03. Although he does not report a long-run wage elasticity, he does obtain an income response to the average of the past four years of R&D that is 0.13 below the short-run income response. From this we will infer that the medium to long-run wage elasticity is approximately 0.10.
Turning to Table 1, we can immediately use this information to calculate an estimate of 9, the elasticity of the private marginal product of R&D with respect to government workers LG, holding the total labor supply constant:
Thus, we may say that in the short-run, at least for countries similar to the United States, the measured nominal private R&D response to an increase in public spending will be positive if the government share is less than about 0.22. If the short-run wage effect were similar in other developed countries, we would expect that countries with larger government shares would display crowding out behavior. This is consistent with the results in Guellec and van Pottelsberghe (1999), who find that countries with high subsidization rates have negative public R&D coefficients in the private R&D investment equation. However, the latter results are based on error-corrected regressions, which deliver long-run rather than short-run behavior. It is also not clear from Guellec and van Pottelsberghe’s paper whether the subsidization measure upon which their analysis focuses really corresponds closely to the measure of public support adopted here – which encompasses all government R&D spending. Reassuring as the reported results are, they have to be taken with a grain of salt. Of course such countries as America and other developed countries have such an opportunity as to provide their citizens with a high quality of life. But there are people who need much money as they get used to earn that’s why they take loans here http://speedy-payday-loans.com/ to satisfy all your needs.
In the case of the United States, the government-financed share of R&D spending fell from about 60 percent to 36 percent over the course of the period examined by Goolsbee (1998). That would imply both that the short-run impact of any increase in public funding – without a shift in the mix favoring grant support – should have been to crowd out private R&D, but that this impact should have weakened appreciably between the late 1960’s and the mid-1990’s. This result is moderately consistent with the trends of reported magnitudes surveyed in David, Hall and Toole (1999), although many of the results pertaining to the U.S. imply complementarity rather than crowding out.
Turning to the long-run response of private R&D, Goolsbee’s estimate of 0.10 for the wage elasticity has the following implication:
We can use this result to bound the long-run elasticity of real and nominal R&D with respect to an increase in government R&D spending. For real spending, the elasticity is
This has two implications: first, the long-run labor supply elasticity of R&D workers needs to be quite large for the real effect to be positive, unless the government share is very small. Second, as the government share of spending rises, it becomes more likely that crowding out would be observed even in the long run.
The elasticity of nominal R&D expenditures, however, is slightly larger than the magnitude just considered (because it includes the positive wage effect, equal to 0.10). The condition for this elasticity to be positive is therefore given by:
which is slightly easier to satisfy. For a country such a Japan, in which the government share of R&D funding was below one-fifth at the beginning of the 1990’s, the critical value of the long-run labor supply elasticity would be something like n > 0.8. It is not implausible, then, that there would have been little crowding out effects from the major program of expanded public funding for science that was called for by the Council for Science and Technology in Japan in 1996 and approved by the government of the day. In view of the thrust of that program towards changing the mix of national R&D expenditures strongly in favor of university-based, “grant” research, it would seem that the plan was particularly well-calculated to yield a long-term positive stimulus to private sector R&D.19 But, as this particular example underscores, to have arrived at the right science and technology policy is not the same thing as having the public sector resources with which to implement it. As a result, one of the less serious consequences of the difficulties that have beset the Japanese economy in recent years has been to deprive us of an interesting natural experiment with which to test the empirical implications of our model.