Academic journal article Contemporary Economic Policy

The Effects of Public Capital on the Growth in Spanish Productivity

Academic journal article Contemporary Economic Policy

The Effects of Public Capital on the Growth in Spanish Productivity

Article excerpt


Productivity growth has slowed down during the last two decades in most Organisation for Economic Co-operation and Development (OECD) countries. This fact has been accompanied by a decline in public capital spending as a share of gross domestic product (GDP). Spain and Portugal are exceptions. To promote long-term equality among countries within the European Union, these countries undertook extensive programs to upgrade their stock of public capital. As a result, gross public capital formation in Spain (as a share of GDP) is currently one of the highest in the EU area (see Table 1). On average, the real value of the country's public capital stock grew by 7.2% during 1986-91, which is considerably above that of most of the OECD countries (Sturm, 1998).

Various authors have tried to determine the effects of public capital on productivity. Starting from the seminal paper by Aschauer (1989), it became increasingly common to believe that a link between public capital and productivity exists. The underlying tenet is that a decrease in the growth rate of public capital spending will be accompanied by a fall in the rate of growth of total productivity. This has generated a vast body of literature, attempting to identify the influence of public investment on the evolution of productivity. (1) Such influence has been traditionally investigated using two different approaches. (2) The production function approach, followed by Aschauer (1989), Ford and Poret (1991), and Munnell (1992) among others, includes public capital as an additional productive factor to the classical inputs of labor and private capital. Assuming a specific production function, generally a Cobb-Douglas with constant returns to scale and a competitive environment, the coefficient accompanying the p ublic capital variables reflects the (direct) effect of public capital on output.

A second procedure, often referred to as the behavioral approach, has been used in Conrad and Seitz (1994), Nadiri and Mamuneas (1994), and Morrison and Schwartz (1996) among others. This consists of deriving a cost function from a firm's dual problem of minimizing costs, where the public capital stock is included as an unpaid fixed input. The translog or the generalized Leontief specifications are usually used to approximate the cost function. From the estimation of such equations, several elasticities, which fully describe the underlying production function, can be computed.

The production function approach has been questioned, and some problems of an econometric nature have been pointed out (endogeneity, nonstationarity, omission of variables, measurement error, etc.). The rigid relationship between public and private capital implied in the Cobb-Douglas form has also been brought into question (see Sturm, 1998, for a detailed discussion of these problems). The imposition of a determined specification in the production function approach may cause bias on the estimates of the impact of public capital, as Berndt and Hanson (1992) have pointed out. On the other hand, the behavioral approach allows for any degree of complementarity or substitutability between fixed and flexible inputs. However, this property also induces the biggest problem; the flexibility of the functional form requires a tremendous amount of information to be included in the database (user costs of private and public capital, intermediate inputs, and its prices, etc.). Furthermore, there may be collinearity proble ms between the regressors, and the estimates are also very sensitive to the specific functional form chosen.

The results of these studies generally asserted the positive effects of public capital, but the range of results was too variable to be conclusive and many deficiencies were subsequently identified due to econometric and specification problems (Pfahler et al., 1 996). (3) Once these problems had been rectified, a number of negative results were published, leading to the conclusion that the positive outcomes of earlier studies were poorly founded (Holtz-Eakin, 1994; Garcia-Mila et al. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.