Author: Alexandre Manoel Angelo da Silva, José Oswaldo Cândido Júnior

Source: Valor Econômico, 09/06/2009

The concern with the macroeconomic effects of public spending (costing and investment) is a topic that encompasses an extensive literature. In this sense, although there is a conventional wisdom that there is absolute superiority of public investment in relation to costing, with regard to a supposed positive impact on the product and a negative impact on it, we affirm that, in general, from the point from a theoretical point of view, there is still no consensus regarding this superiority, neither in the short nor in the long term. In fact, depending on the way the production function is modeled, positive or negative performances are obtained from public investment or costing on the product.

By way of illustration, renowned authors in the economic growth literature, such as Robert Barro, consider some specific types of costing expenditure to be productive. In particular, those goods provided publicly subject to congestion (expenses with maintenance of water supply and sanitary sewage services, in the Judiciary and public security) and expenses that contribute to the accumulation of human capital (expenses with education and training). Note that, in the concept of national accounts, the majority of these expenditures are allocated as consumption (costing) of the government and could (by conventional wisdom) have their increase understood as harmful to economic growth, when, according to Barro's modeling, this the same increase would positively affect the productivity of the factors of production.

In turn, there are authors, such as Lant Pritchett, who point to a lot of empirical evidence of investments that are considered “white elephants”, that is, public investments that do not expand the productive capacity of the economy. There are also authors such as Devarajan, Swaroop & Heng-fu who highlight the importance of the marginal effects of public spending, indicating that a sharp expansion of a certain type of expenditure considered productive may render it unproductive, that is, the productivity of the spending depends on its level and composition that affect its marginal effects.

Given this non-consensual theoretical framework regarding the actual impacts of public spending (costing and investment) on the product, it is natural that many economists are skeptical and prefer to empirically assess the macroeconomic effects of public spending, without attributing a priori a superiority to public investments. . Among so many other strategies by economists who have worked on this theme around the world, this was the one adopted by Perotti (Perotti, R. (2004). “Public Investment: Another (Different) Look”. University of Bocconi) and Mittnik & Neumann (Mittnik, S. and T. Neumann. (2001). “Dynamic Effects of Public Investment: Vector Autoregressive Evidence from Six Industrialized Countries”, Empirical Economics 26, 429-446). It was also this strategy that we adopted in the Discussion Text to be published by Ipea, under the title “Macroeconomic Impacts of Public Spending in Latin America”.

In our case, we assessed the macroeconomic impacts of the main components of public spending (consumption and investment) on GDP, household consumption and private investment, in a sample of six countries in Latin America. The countries chosen were Argentina, Brazil, Chile, Colombia, Mexico and Venezuela in the period 1970-2003. This is an eminently exploratory study (of data) from an econometric point of view, which formally does not test any theoretical model.

In the specific case of Brazil, in the long run, we found a negative relationship between government consumption (costing) and the product and, due to the absence of cointegration, it was not possible to observe the relationship between public investment and product. However, in the short term, we did not obtain superiority in public investment, so it is irrelevant for the government to use public funding or investment to recover economic activity in the short term, contrary to what conventional wisdom points out.

It should also be noted that, when analyzing the results of all the countries investigated in our research, we conclude that the effect of costing on the product depends on the level and how the costing evolves at the margin. In other words, the fiscal space allowed to increase spending costs seems to depend on how the following relationship evolves: level of government consumption over public investment. In countries where this ratio is high, this fiscal space is smaller, and vice versa. Thus, Argentina and Brazil have a small fiscal space to expand public consumption (costing) in relation to Chile and Mexico, for example.

Naturally, in the case of Brazil, this smaller fiscal space stems from the notorious fact that, in the last decade, the fiscal adjustment basically fell on public investments, preserving current expenses, which, in most cases, are protected by legal provisions and constitutional requirements. Thus, suggesting a switch from costing to investment is obviously easier to recommend than to implement. In fact, this is a discussion that goes beyond the ideological sphere, being, therefore, entangled in our institutional structure, which does not contain an adequate incentive scheme to seek a “great” composition of public spending.

In conclusion, it cannot be ruled out the hypothesis (apparently not testable) that certain public investments, such as those in infrastructure, generate positive expectations in the economy, carrying new private investments in the medium and long term, although they do not produce significant effects on the product in the short term. In this way, the fiscal space for these investments would be justified as a contribution of fiscal policy to a "more" consistent recovery in economic activity, especially in this environment of deteriorating expectations.

Alexandre Manoel Angelo da Silva is a researcher at Ipea.

José Oswaldo Cândido Júnior is a researcher at Ipea assigned to the Senate.