Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
An analysis of the impact of public infrastructure on productivity performance of Mexican industries. E.C. Mamatzakis* June 2007 Abstract It has been frequently quoted in the literature that one decisive cause of the productive performance of an economy might be public infrastructure. This paper provides a simple theoretical specification of alternative measures for accounting the effects of public infrastructure on economic performance in terms of gains in profits, cost savings as well as in terms of productivity growth enhancement. In an empirical application we opt for Mexican industry data. The results show that infrastructure capital asserts significant profit gains, though some variability across time exists. This is due to the fact that potential growth is constrained by low levels of public fixed capital formation, and thus in the case of Mexico physical infrastructure is inadequate. JEL Codes: D21, D24, H54, E62 Keywords: Dual profit function, productivity growth, public infrastructure, Mexican manufacturing. *Emmanuel Mamatzakis, Department of Economics, University of Macedonia, Egnatia 156 Thessaloniki 540 06, Greece, e-mail: [email protected]. 1 1. Introduction Measuring productivity growth has always been central in discussions regarding the development of the Mexican economy (Feltenstein and Ha, 1999). Jorgenson (1997) defines productivity growth as “the part of output growth that can not be explained by an increase in the use of inputs”. Moreover, productivity growth is attributed to improvements in technology, scale effects and an increase in the efficiency of resource use (see Capalbo and Antle, 1988). However, given the complexities involved in accurately measuring productivity growth it is not surprising that much controversy has been generated around this issue. A rather neglected determinant of productivity growth for a prolonged period of time is public infrastructure, though its importance has been unequivocal. The spark of recent research papers could be traced to Aschauer (1989) following the early work of Meade (1952). Ashauer’s paper argued that infrastructure explained some of the productivity growth slowdown in US economy in the late seventies and it triggered a plethora of papers thereafter (for a review see Gramlich, 1994 and Vijverberg et al., 1997). Despite the evidence provided by Aschauer (1989) and Munnel (1992), some research provided estimates of an insignificant return to public infrastructure in US (Evans and Karras, 1994 and Holtz-Eakin, 1994). Moreover, the high output elasticities of infrastructure reported by Aschauer (1989) and Munnel (1992) raised criticism on issues such as the lack of flexibility of their production function specification, the underlying aggregation bias in the macroeconomic data sets, and the possible endogeneity of output (see Vijverberg et al. (1997)). Using duality theory, Nadiri and Mamouneas (1994) and Morrisson and Schwartz (1996) addressed these issues to find that public infrastructure was enhancing productivity in US. 2 Despite the numerous studies on the return to public infrastructure, few studies have attempted to measure this return in the case of Mexican economy. A country in which the role of public infrastructure may be seen as particularly influential is Mexico, given the government reduction of public investment from 12% in the early eighties to bellow 5% in the early nineties. This development comes in parallel with a sluggish economic growth performance and a major economic crisis in the mid nineties. Moreover, OECD (2003) argue that Mexican industry has been under stressed due to the restructuring process triggered by the intensified competition of low labour cost countries, such as China. This process has highlighted the detrimental impact stemming from chronic low investment. Indeed, in the low skilled manufacturing sector, Mexico is not able to compete against China or with other low income countries, including in Central America. Based on data reported in the IMD World Competitiveness Yearbook (2004) the hourly compensation in the manufacturing sector in Mexico is $2.45 compared to $0.66 in China. In addition, during the nineties the Mexican economy faced hefty macroeconomic imbalances due to the banking crisis in 1995, which coupled with rising world uncertainties posed by high volatility in oil prices and high interest rates, did not benefit industry. Besides the idiosyncratic characteristics of Mexican industry and the uncertainties linked to the external economic environment, some additional important economic factors, such as globalization, have also contributed to the economic performance of the sector. Moreover, globalization has changed the way firms operate, and where they choose to locate. In addition to the effect of the above factors on Mexican economic performance, infrastructure investment could contribute to total factor productivity (TFP), regardless of technical change and returns to scale effects. Earlier research using dual cost function framework indicates that public infrastructure may indeed be one of the productive inputs for Mexico (see Shah, 1992, Feltenstein and Shah, 1995, Feltenstein and Ha, 1999). 3 The analysis of this paper complements the above studies for the Mexican industry as it also opts for a flexible functional form, and therefore it departs from the primal analysis proposed by Aschauer (1989). Moreover, we derive a solution to the profit maximisation problem that a firm is facing. In turn, this optimization provides a theoretical framework that allows the identification of profit gains due to public infrastructure. In addition, this paper provides a theoretical specification of profit function that would allow measuring the effects of infrastructure of total factor productivity, as well as in terms of cost savings. The main advantage of this theoretical specification is its ability to provide a single framework that allows quantifying the productive performance of infrastructure capital. The choice of profit function is based on the earlier research of Vijverberg et al. (1997), arguing that the profit function approach in general performs better than either the production function or cost function approach (see also Shah, 1992). The remainder of the paper is organised as follows; section 2 presents the theoretical framework of the profit function, while section 3 discusses the data set and section 4 the empirical specification. Section 5 provides the estimation procedure, and the empirical findings, whereas the last section highlights some concluding remarks and economic policy implications derived from the empirical findings. 2. Theoretical specification Consider the following production function, where X, G, t denotes production inputs, public infrastructure, and technological change respectively. 4 Y = f(X, G, t) (1) The firm’s objective is to maximise profits given the production function (1) and it can be written as: ʌʌ(P,w, G, t) = max [P f(X, G, t)-w X] (2) , where P is the output price, w is as nx1 vector of the price of private inputs. The profit function is strictly convex in P and w. By applying the envelope theorem we get: ʌʌG(P,w, G, t) = P fG(X, G, t) (3) ʌʌt(P,w, G, t) = P ft (X, G, t) (4) , where subscribes describe first derivatives. Another way of looking at the same optimisation is to maximise the difference between total revenues and the cost of producing the output level Y as: ʌc (P,w, G, t) = max [P Y – C(w, Y, G, t)] (5) as above we can apply envelope theorem and obtain : 5 ʌcG (P, w,G, t) = -CG (w, Y, G, t) (6) ʌct (P,w, G, t) = - Ct (w, Y, G, t) (7) Equation (6) depicts that the marginal shadow value of public infrastructure as measured from the profit function is equal to the negative of the marginal shadow value of public infrastructure as measured from the cost function C. Similarly, equation (7) depicts for the technological change. The above equations can measure the effects of public infrastructure on private economic performance from the perspective of profit function. Profit gains due to public infrastructure In the case that infrastructure capital is productive input, then infrastructure would induce profit gains. The effect of public infrastructure on profit is provided as ȘʌG: . . . ȘʌG = ( P, w, G, t ) ( P,W ) (8) , where dots above the variables denote growth rate and ȟ is a function of the growth rate of P and W. For practical reasons ȟ is taken as the weighted average of the growth rates of output and input prices with weights being the elasticities of the profit function with respect to P and W. Thus, we derive: 6 . P ( P, w, G, t ) P . Wi ( P, w, G, t ) wi . ȘʌG = ( P, w, G, t ) P wi ( P, w, G, t ) ( P, w, G, t ) (9) Now, by total differentiating the profit function we get the growth rate as: . ( P, w, G, t ) P ( P, w, G, t ) P . wi ( P, w, G, t ) wi . G ( P, w, G, t )G . t ( P, w, G, t ) P wi G ( P, w, G , t ) ( P, w, G, t ) ( P, w, G, t ) ( P, w, G, t ) (10) Combining equations (9) and (10) with (8) we get: ȘʌG = G ( P, w, G, t )G . t ( P, w, G, t ) G ( P, w, G, t ) ( P, w, G, t ) (11) In effect the ȘʌG measures the impact of public infrastructure and technological change on profit over time (see Ray and Sagerson, 1990, and Fousekis and Pantzios, 2000). Cost savings due to public infrastructure Similarly, in a parallel exercise we can derive the cost saving impact of public infrastructure as in Morrison and Schwartz (1994) using cost elasticities with respect to input prices as weights: . ȘCG = Y CG ( w, Y , G, t )G . Ct ( w, Y , G, t ) G C ( w, Y , G, t ) C ( w, Y , G, t ) (12) . C , the cost elasticity with respect to output. , where C 7 Given the assumption of cost minimization, the equation (12) decomposes the cost savings into the scale effect, the effect of public infrastructure, and the technical change effect. Moreover, in order to capture the effect of public infrastructure on productivity growth we have to take into account that both price changes as well as changes in public infrastructure and technology affect output. Note, if the supply function is given as Y = f (P,W, G, t), then the growth rate of output adjusted for possible endogeneity within the profit function framework gives: . . YP ( P, w, G, t ) P . YWi ( P, w, G, t ) wi . ( P, w, G, t ) P wi Y ( P, w, G, t ) Y ( P, w, G, t ) (13) , where subscript Į counts for the adjusted profit maximized growth of output. Knowing that the growth rate of Y is given as: . ( P, w, G, t ) = YP ( P, w, G, t ) P . YWi ( P, w, G, t ) wi . YG ( P, w, G, t )G . Yt ( P, w, G, t ) (14) wi G P Y ( P, w, G, t ) Y ( P, w, G, t ) Y ( P, w, G, t ) Y ( P, w, G, t ) By combining (13) and (14) we get: . YG ( P, w, G, t )G . Yt ( P, w, G, t ) G Y ( P, w, G, t ) Y ( P, w, G, t ) (15) 8 Thus, the adjusted, under the assumption of profit maximization, cost saving is: . ȘCGĮ = Y CG ( w, Y , G, t )G . Ct ( w, Y , G, t ) G C ( w, Y , G, t ) C ( w, Y , G, t ) (16) The cost saving rate is decomposed: (a) into the scale effect induced by the response of production to changes both in public infrastructure and technology, (b) into the direct cost impact of public infrastructure, that is the contribution of public infrastructure to the firm’s cost savings over time holding production constant, and (c) the dual technical change effect. Now, by substituting equations (6) and (7) into (16), multiplying and dividing the lat two terms on the right hand side of equation (16) by profit, and knowing that ʌ/c = (R-C)/C =ı-1, where R is total revenue, we get: . ȘCGĮ = (ı-1) (-ȘʌG ) + ı Y (17) The impact of public infrastructure into productivity growth The overall productivity impact of public infrastructure, as in Vijverbeg et al. (1997), given the production function 1 is given as: . TFP G = fG ( X , G, t )G . ft ( X , G, t ) G f ( X , G, t ) f ( X , G, t ) (18) Now, substituting (3) and (4) into (18), and multiplying and dividing the right hand side of (18) by profit we get: 9 . TFP G = 1 ȘG (19) 3. Data set for Mexican economy At the outset it is worth mentioning that the case of Mexico is of some interest as investment in core infrastructure, that is in electricity, transport and communication, has been one of the lowest among OECD countries (see Diagram 1), despite some improvement over the a period of twenty years is observed. There many causes of underinvestment in infrastructure capital. OECD (2005) argues that one of the most important one is closely linked with lack of fiscal consolidation and prioritisation of public expenditure towards investment rather than consumption expenditure. For example, public investment was severely curtailed for budget consolidation purposes following the 1995 peso crisis, whilst in addition the absence of a coherent public investment strategy meant that investment projects were crucially depending on changes in oil revenue. Moreover, the economic recovery in the second half of the nineties did not bring benefits in terms of raising potential growth, as the investment boom was largely focused on consumption related facilities, such as shopping malls and fast food chains, rather than production. Based on the country economic review of OECD (2005), the inadequate public investment has created shortages in basic infrastructure such as communication, transportation, electricity, sanitation and water.1 In addition, the business climate in Mexico is not at all supportive to private investment initiatives as a complementary provision of the needed infrastructure services. This is so due to 1 According to OECD Environmental Performance Review (2003) in Mexico the water and waste water sector would require $2.2 billion of investment funds, twice the annual budget of the National Water Commission (CNA), which is responsible for producing and regulating water. 10 heavy regulatory and legal restrictions, which it is striking that these are part of the institutional infrastructure that also appears to be inadequate to enhance potential growth. Diagram 1, Aggregate infrastructure indicators in OECD countries, USA 1995 = 100 2000 1980 120 120 100 100 80 80 60 60 40 40 20 20 TUR POL MEX HUN CZE PRT KOR ITA DEU ESP GRC JPN NLD BEL GBR CAN NZL FRA AUT AUS CHE ISL FIN DNK IRL USA SWE 0 NOR 0 Source: Nicoletti et al. (2003). In turn an inadequate provision of infrastructure deters further investment and act as an impediment to business. In particular, weaknesses in transport and communication infrastructure prevent Mexico from getting the most out of its proximity to the US. Diagram 2 highlights that besides the strategic location of Mexico fixed investment as percent of GDP in Mexico takes low values if compared to other OECD countries. The investment ratio averaged merely 20% in the latest expansion phase during the period 1996-2001, including residential construction and investment by large state-owned companies. This ratio is lower than its level in the early eighties or in previous decades, and it is not high by OECD standards. Following the extensive privatisation operations of the early 1990s, the public sector share of investment declined. There is also evidence of inadequate quality of investment (see OECD, 2005). 11 Diagram 2, Fixed investment ratios (% of GDP), in nominal terms, average 1996-2003 35 35 30 30 25 25 20 20 15 15 10 10 KOR SVK JPN CZE PRT ESP AUS IRE HUN CHE GRC AUT ISL LUX NLD POL TUR NZL DEU BEL NOR DNK CAN MEX FRA USA ITA 0 FIN 0 GBR 5 SWE 5 Source: OECD 2005, Economic Outlook, database. The above descriptive analysis shows the significance of applying the theoretical specification of the profit function using Mexican data. Thus, in an empirical application we use data from Mexican industries over the period 1970-2000. The data set is mainly derived from the Annual Industrial Survey (AIS) from the Mexican Institute for Statistics, Geography and Informatics (INEGI), which provides information in manufacturing regarding the following aspects: output, employment, investment, capital stocks, and expenditures in electricity, communications, and transport.2 We focus in Mexican industry to employ some disaggregation into our empirical application, thus ten Mexican industries are opted. Those industries 2 We would like to thank Professor Jose Miguel Albala-Bertrand and Professor Salgado Banda Héctor for providing the source of much of the data set. 12 are: mining, food, beverages & tobacco, wood and wood products, paper, chemicals, plastic & rubber, metal products, machinery & equipment, construction. The series for infrastructure capital is constructed using several series for total Gross Fixed Capital Formation (GFCF) and investment. The capital stock series for both total capital and infrastructure components were built up via a Perpetual Inventory Method (PIM) applied to a benchmark capital stock for the year 1970, which is the standard OECD method. A PIM is a method that adds GFCF to a benchmark capital and subtracts the depreciation due in each year. The depreciation pattern can be linear or not. We used a linear depreciation pattern, which is the normal recommendation when information about actual depreciation is scattered and/or unreliable, as is in our case (AlbalaBertrand, 2003). The benchmark for total capital stock was based on Hofman (2000a) and Hofman (2000b), while the proportion of core infrastructure in that stock is based on the methodology proposed by Arellano & Braun (1999). In turn, the proportion of infrastructure components in total infrastructure was based on their actual investment patterns. The depreciation rates used were the ones suggested in these sources. The price indexes to deflate nominal series mostly came from the GDP deflator, but we also used the PPI and CPI when the former was unavailable. All series are available in 1993 constant pesos in the NIPA statistics. 4. Empirical model The cost function of Mexican industries is assumed to be affected by the provision infrastructure capital. As a result the profit function takes the form ʌ = ʌ(P, wi, G, t) (20) 13 , where P is the price of output, wi are n dimensional vector of input prices for labour and capital, G is infrastructure capital, and t is an index of technological change. To estimate the effects of infrastructure capitals on the productivity and production structure of the Mexican industries, we specify a tranlog restricted profit function: lnʌ = Į0 + n 1 n 2 i 1 Įi lnwi + plnP + t t + G lnG + i 1 n + Ȗip lnwilnP + i 1 n m 1 1 GG lnG2 + pplnP2 2 2 aij ln wi ln wj + j 1 iG ln w ln G + pGlnP lnG + ȕtt + i i 1 1 tt lnt2 + 2 n Ȗit lnwit + pt lnPt + i 1 Gt lnGt (21) Using Hotelling’s Lemma to equation (21), we obtain the following equations for output, input, and shadow infrastructure profit share: RP = Si = RG = d ln = p + PPlnP + d ln P d ln = Įi + d ln wi n iP ln wi + PGlnG +Pt t (22) i 1 n n if ln w + j i 1 d ln = G + GGlnG + d ln G ĮijlnPi + i 1 n n i 1 n ȖiGlnG + Ȗitt ln w + pGlnP +ȖGtt iG (23) i 1 i (24) i 1 The motonicity condition on the profit function requires that it is non decreasing (non increasing) in prices for all outputs (inputs) and non-decreasing in infrastructure capital. At the point of approximation, these imply that the output and shadow (input) profit shares are positive (negative). Sufficient conditions for these are that ȕp0, ȕt0, and Įi 0 for all i, respectively. We also impose restrictions associated with linear homogeneity in the profit function (equation 21) with respect to n output and input prices as i +p =1, i 1 n it i 1 +Ȗpt =0, n n + ij i 1 Ȗip =0. For symmetry, we also i 1 14 impose Įij=aji. In addition, convexity with respect to price is tested using the Hessian matrix of second order derivatives that needs to be positive semi-definite. Also, the profit function needs to be concave with respect to the quasi fixed inputs, in this paper the capital infrastructure stock, and thus the associated Hessian matrix should be negative semi-definite. Empirical results The system of equations (21), (22), and (23) is estimated with iterative SUR to account for contemporaneous correlation of error terms, while we exclude the intermediate input’s profit share equation to avoid singularity of the variance co-variance matrix. Also, in order to estimate the parameters of the profit function equation we pooled the inter industry time series data. By doing so we deal with the problem of multicolinearity frequently associated with data of a single industry. The interindustry data set provides the necessary variability, and therefore allows a more rigorous statistical analysis of the parameter estimates and the correspondent elasticities. As a way to capture these interindustry differences we introduce dummy variables on the constant and slope of the cost function for each industry. We have assumed that Į0s = Į0 + Ds , where Ds 0s s refers to the industry dummies taking values 1 and 0, s is the industry identification index, and the js are normalised with respect to the k industry (jk = 0). We could also introduce industry dummies for the cross terms and infrastructure capital; we opt not to for the simplicity of the current empirical model. Parameter estimates of the profit function are reported in Table 1. Overall, the results suggest that the estimated translog profit function is well behaved, as the signs on the coefficients of the profit 15 function are reported to be consistent with curvature conditions, while the magnitudes of the estimated elasticities are plausible and statistical significant for most industries. Moreover, the fitted profit function satisfies the motonicity property at all data points as the output shares are found to be positive and the variable input shares negative, while the profit shares of infrastructure capital is estimated to be positive. Also the Hessian matrix reports that the profit function is convex in prices and concave in infrastructure capital.3 Table 1: parameter estimates of translog profit function. Estimated Value Standard Error Parameter ĮK ĮL ȕP ȕPP ȕG LG ȖPK ȖPL PP ĮKL ĮKȀ ȕt tt GG tP tL tK Gt 1.75 -1.37 -0.31 1.62 0.59 0.38 0.05 1.52 0.84 1.36 -1.45 -0.39 0.0064 0.002 -0.65 0.031 -0.032 0.007 0.049 2.74 2.05 2.22 2.87 3.44 2.67 2.79 2.92 2.27 2.13 -2.35 -1.92 2.71 2.79 -2.28 2.44 -2.03 2.17 2.013 3 In order to take into account the major macroeconomic instability in the mid nineties a dummy is included in the estimation of the translog profit function. It is found to be significant. 16 Diagram 2 presents the elasticity of profit with respect to public infrastructure. It is positive across all sample points. This finding implies that public infrastructure asserts a positive externality to the Mexican industry. However, note that it follows a negative trend till mid nineties, whereas in 1995 the financial crisis led to major macroeconomic instability that resulted to a negative spill over in the impact of infrastructure on profits. Diagram 2, The elasticity of profit with respect to infrastructure 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 19 70 19 72 19 74 19 76 19 78 19 80 19 82 19 84 19 86 19 88 19 90 19 92 19 94 19 96 19 98 20 00 0 The effect of public infrastructure on profit Note that the profit gains of public infrastructure depend crucially on the elasticity of profits with respect to public infrastructure, but also the actual growth rate of public infrastructure. Despite the promising seventies in terms of investment in infrastructure, that averaged around 10%, Diagrams 3 depicts a negative trend from early eighties till the crush in 1995 due to the financial crisis. Moreover, the average growth rate of infrastructure capital was 5.5% in the eighties, almost half the one of the seventies, while it dropped to -7.5% in the mid nineties. A recovery in infrastructure 17 investment is reported in the late nineties, reaching 8.5% in 2000, as the outcome of an improvement in the general economic climate. Diagram 3, growth rate of infrastructure capital 0.2 0.1 0 19 71 19 73 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 -0.1 -0.2 -0.3 -0.4 -0.5 -0.6 -0.7 -0.8 -0.9 Table 2 presents the contribution of public infrastructure to profit as derived from equation (11), augmented with the technical change effect, over the sample period 1970-2000. The results show that the effect of infrastructure is positive in all years but in the mid nineties. The average value of the profit gains over the period is around 0.74%, suggesting that a 1 percent increase in public infrastructure has, on average, raised the level of profits in Mexican industries by 0.74%. The technical change in equation (11) is also positive every year but 1995, insinuating that technical change was progressive with an average value of 0.25%. This suggests that the impact of public infrastructure is more than double the impact of technical change. Alas despite the magnitude of profit gains due to infrastructure, over time the ȘʌG exhibits a clear downward trend. The average ȘʌG during the period 1970-75 is around 0.15%, whereas some decline is observed in the period 1976-80, 18 followed by a sharp decline thereafter to reach an all time low at around -0.14% in the first half of the nineties. This development is explained by both the decline in the profit elasticity with respect to public infrastructure but also the downward trend observed in the growth rate of public infrastructure since the eighties and till mid nineties, and in particular due to the crush of investment in infrastructure investment in 1995. TABLE 2, estimate of rate of gains in profit due to infrastructure The cost savings effect of public infrastructure Based on equation (17), Table 3 presents the cost savings impact of infrastructure capital, ȘCG, that is the direct effect of public infrastructure on cost savings. The average ȘCG is -0.209 over the sample 19 period, implying that public infrastructure is a productive input for Mexican industry. As in the case of ȘʌG, the ȘCG exhibits also a downward trend, from around -0.3in the seventies to -0.19 during 1980-85, further down to -0.11 in 1886-90 to reach the lowest values during the period of the banking crisis in the 1991-95, -0.10. Some recovery is reported in the late nineties. TABLE 3, cost savings due to infrastructure The effect of public infrastructure on TFP The effect of public infrastructure on TFP and its decomposition into the direct impact of public infrastructure and the primal rate of technical change are obtained from equation (18). Over the 20 sample period TFP growth followed a downward trend, falling from 1.21 (relative TFP, US=1) in 1970 to 0.67 in 1995, while partly recovering thereafter to reach 074 in 2000. It turns out that the average rate of TFP growth is lagging behind that of US, and most importantly emphasises the magnitude of catching up for the Mexican economy if it was to converge to the living standards of its northern neighbor. Diagram 4, the impact of infrastructure on productivity growth 0.160 0.140 0.120 0.100 0.080 0.060 0.040 0.020 19 71 19 73 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 0.000 As it is the case with the ȘʌG and the ȘCG the contribution of public infrastructure of TFP exhibits a downward trend (see Diagram 4). The average the contribution of public infrastructure effect on TFP growth is 0.092. It falls from around 0.14 in the seventies to 0.087 in 1981-85, 0.052 in 1986-90, to a further decline of 0.034 in the early nineties. This downward trend in the TFPG is the joint outcome of both the dramatic decline in the growth rate of public infrastructure as well as the decline in profit gains due to infrastructure capital shortages. 21 TABLE 4, average values of the effect of public infrastructure on productivity Growth 6. CONCLUSION In most respects, Mexico’s economic performance improved in the late nineties. However, Mexico’s growth performance since the restoration of macroeconomic stability has been rather unsatisfactory as potential GDP growth estimates have been revised down to below 4% (OECD, 2005). This performance is quite slow for a country with low levels of income and productivity and high rates of population growth, whereas it is too slow to narrow the gap in living standards relative to other OECD countries. In addition, labour productivity remains on a declining path as it fell from 0.5% before the financial crisis to 0.1% in 2000. This paper develops a theoretical framework based on a restricted profit function that provides evidence of the positive contribution of public infrastructure on the returns of public infrastructure in terms of higher (lower) profits (costs) for Mexican industry. In addition, a decomposition of productivity growth shows that infrastructure investment may well be responsible for the observed slow down since early in the eighties. This suggests that productivity growth cannot be attributed to technical change and scale economies alone. Overall, is shown that public infrastructure increases (reduces) the total profit (cost) of Mexican industries and, thus, can be regarded as a productive input that contributes to productivity growth. 22 The results of the present paper emphasise the necessity to address the shortages in core infrastructures, such as roads, water supply and water treatment. Moreover, improving the transportation and electricity system would raise productivity growth. However, an issue that the current paper has not tackled concerns the issue of raising the appropriate financial resources to build infrastructure. OECD (2005) argues that for Mexico “fiscal revenues under the existing tax system are insufficient to finance infrastructure spending by federal, state and local governments at an adequate level”. In addition, it is more than often the case that fiscal consolidation efforts weigh much on the public investment rather than on the public consumption expenditures. Involving the private sector in the provision of some public services could be part of the answer to raise infrastructure capital, raising tax revenues and directing oil revenues to public investment could also play a positive role. 23 References Albala-Bertrand, J.M. & Mamatzakis, E.C. (2004) “The Impact of Infrastructures on the Productivity of the Chilean Economy”, Review of Development Economics, Vol. 8, Issue 2, pp. 266-78. Albala-Bertrand, J.M. (2003) “An Economical Approach to Estimate a Benchmark Capital Stock. An Optimal Consistency Method”, QMW Working Paper No. 503. Arellano, M.S. & Braun, M. (1999) “Stock de Recursos de la Economia Chilena”, Cuadernos de Economia, No 107, pp. 639-84. Aschauer, D.A. (1989) “Is Public Investment Productive”, Journal of Monetary Economics, 23, 177200. Bacha, E. (1990) “A Three-Gap Model of Foreign Transfers and the GDP Growth Rate of Developing Countries”, Journal of Development Economics, 32, 279-296. Barro, R. (1989) “A Cross-Country Study of Growth, Savings, and Government”, NBER Working Papers No. 2855. Blejer, M. and Khan, M. (1984) “Government Policy and Private Investment in Developing Countries”, IMF Staff Papers, Vol.31, No.2, 379-403. Chenery, H. (1953) “The Application of Investment Criteria”, Quarterly Journal of Economics, No. 57, 123-140. Christensen, L., Jorgenson, D., Lau, L., 1971. Conjugate Duality and the Transcendental Logarithmic Production Function. Econometrica, 39, pp. 28-45. Coeymans, J.E. “Determinantes de la Productividad en Chile”, Cuadernos de Economia, No 107, pp. 597-638. Demetriades, P., and Mamuneas, T., 2000. Intertemporal Output and Employment Effects of Public Infrastructure Capital: Evidence from 12 OECD Economies, The Economic Journal, Vol. 110, No. 465, pp. 687-712. Feltenstein, A. and Ha, J. (1995) “The Role of Infrastructure in Mexican Economic Reform”, The World Bank Economic Review,Vol. 9, No. 2, pp. 287-304. Feltenstein, A. and Shah A. (1995), “General Equilibrium Effects of Investment Incentives in Mexico”, Journal of Development Economics, 46: 253-69. Fousekis P. and Pantzios C. (2000), “Public Infrastructure and Private Economic Performance: Alternative Measures with an Application to Greek Agriculture”, Revista Italiana de Economia, 47, 1, pp.118-128. Gramlich E. M (1994), "Infrastructure Investment: a Review Essay", vol. XXXII, Journal of Economic Literature, Sept., 1176-96. Hirschman, A.O. (1958) The Strategy of Economic Development (New Haven, Yale University Press, New Haven). Hofman, A. (2000a) The Economic Development of Latin America in the Twentieth Century (Cheltenham, Edward Elgar). Hofman, A. (2000b), “Standardised Capital Stock Estimates in Latin America: a 1950-94 Update”, Cambridge Journal of Economics, 24, pp. 45-86. Holtz-Eakin, D., 1994. Public Sector Capital and the Productivity Puzzle. Review of Economics and Statistics, Feb., 76(1), pp.12-21. Hulten, C.R., and R.M. Schwab. 1991. “Public Capital Formation and the Growth of Regional Manufacturing Industries.” National Tax Journal. 44(4):121-134. IMD, 2004, World Competitiveness Yearbook. 24 Judge, 1980. The Theory and Practise of Econometrics. Wiley, New York. Nadiri, M.I., Mamouneas, T.P., 1994. The Effects of Public Infrastructure and R&D Capital on the Cost Structure and Performance of U.S. Manufacturing Industries. The Review of Economics and Statistics, 76(1) pp.22-37. Morrison, C. J. and A. Schwartz, 1994. “Distinguishing External from Internal Scale Effects: The Case of Public Infrastructure,” Journal of Productivity Analysis, 5(3), pp.249-70. Morrison, C. J. and A. Schwartz, 1996. State Infrastructure and Productive Performance. The American Economic Review, pp 1094-1111. Munnell, A. (1992) “Infrastructure and Economic Growth”, Economic Perspectives, Vol. 6, No. 4, pp.77-96 Nadiri, M.I. and Mamuneas, T.P. (1998), “The Effects of Public Infrastructure and R&D Capital on the Cost Structure and Performance of U.S. Manufacturing Industries,” Review of Economics and Statistics, 76(1), pp.22-37. Nicoletti G. et al. (2003), “Policy Influences on Foreign Direct Investment,” OECD Economic Studies, June. Nurkse, P. (1954) “International Investment Today in the Light of Nineteen Century Experience”, Economic Journal, No. 64, pp.45-67. Ortiz, G. and Noriega, C. (1988) Investment and Growth in Latin America (IMF, mimeo). OECD, 2003 “EDRC Review of Mexican Economy”, Paris. OECD, 2005 “EDRC Review of Mexican Economy”, Paris. OECD, 2003, “Environmental Performance Review: Mexico”, Paris. OECD, “Economic Outlook”, various years, Paris. Ray S.C. and Segerson K. (1990) “A Profit Function to Measuring Productivity Growth: the Case of US Manufacturing”, Journal of Productivity Analysis, 2, pp. 39-52. Shah, A. (1992) “Dynamics of Public Infrastructure, Industrial Productivity and Profitability.” The Review of Economics and Statistics, 74(1): 28-36. Sturm, J.-E. 1998. Public capital expenditure in OECD countries: the causes and impact of the decline in public capital spending. Edward Elgar Publishing Limited. Cheltenham. Taylor, L. (1991) Income Distribution, Inflation, and Growth (Cambridge, Mass., MIT Press). Taylor, L. (1983) Structuralist Macroeconomics (New York, Basic Books). Vijverberg W., Vijverberg C. and Gample J., (1997) “Public Capital and Private Productivity”, Review of Economics and Statistics, 79, pp. 267-78. 25