Total Factor Productivity in Food Industries of Iran

Considering to Importance of Food Industries ( Priority of the Non-oil Exports in Foreign Trade, respond to Nutrition of population and Prevention of Wastage) This paper examine the levels of labor, total productivity and technical changes in food industries and compare with total industries of Iran over the 1971-2006 period. The results show that labor productivity and total factor productivity in food industries were lower than average total industries over the period. Also the estimation of technical changes show that measure of technical change in food industries was 0.09 percent while for total industries was 0.16 percent over the period.

International organizations of productivity (APO & OECD) (Note 2) have attempted to present a practical guide for the measurement of productivity. Their attempt has been used to compare economies in terms of productivity. In recent years, there has been an increasing interest in the examination of productivity from different parts of the economy such as industry, agriculture, and services. Numerous studies have attempted to explain productivity in the economic sector, for example, productivity growth in Swedish manufacturing (Carlsson, 1981), the impact of regional investment incentives on employment and productivity in Canada (Daly, Gorman, Lenjosek, MacNevin, & Phiriyapreunt, 1993), productivity and imperfect competition in Italian firms (Contini, Revelli, & Cuneo, 1992), explaining total factor productivity differentials in urban manufacturing of U.S (Mullen & Williams, 1990). Total factor productivity growth in manufacturing has been examined by applied parametric and non-parametric approaches. In most of the studies have used non-parametric approach, wherein total factor productivity growth has decomposed into efficiency change and technological change. Efficiency change measures ''catching-up'' to the isoquant while technological change measures shifts in the isoquant. For example,see Weber and Domazlicky (1999) ;Nemoto and Goto (2005); Yu (2007); (Maniadakis and Thanassoulis (2004) and Radam (2007). Several researchers used econometric approaches to estimate the level of TFP and growth rate of TFP in manufacturing. In this approach, the growth rate of TFP is measured as the residual growth in value added in manufacturing, after accounting for the contribution of input growth to value added. Lach (1995), Windle and Dresner (1992), Rushdi (2000), Eslava et al (2004), Lam and Lam (2005)and Mollick and Cabral (2009). In these researches, Translog production function and Cobb-Douglas production function form have been applied to estimate TFP growth and estimate the share of production inputs that utilized in index method. In recent years, several attempts have been made to investigate productivity in different sectors of Iran economics. Most studies in productivity have only been carried out in a sectoral or regional areas of economy, for example; Salimifar (2005) utilized translog production function for computing total factor production growth in Khorasan province industry of Iran. The scholar applied Kendrik index for accounting total factor production level. Askari et al. (2007) utilized the primer index, Solow index, Kendrik index, Divisia index and Tornvist index to investigate productivity in rural industries of Iran. The intellectuals compared all the production function (Cob-Douglass, Debertin, Translog and CES) and found that the Cob-Douglass production function was suitable for the industry. Bakhshali and Mojtahed (2005) carried out a comparative investigation of technology change on productivity of inputs in the industrial and agricultural sectors. The scholars utilized Cobb-Douglas production function to obtain technology change on productivity and found that the effects of technology change in industrial sector was more than the agricultural sector, technology change for industry and agricultural were 0.04 and 0.03 respectively. However, given the volume of works done in other countries on the concept of productivity, much work still needs to be done within the Iranian context. So far these studies have only been applied to investigate productivity in total industries. On the other hand, lack of research related to productivity in food industry of Iran has existed as a problem for many years. To fill the existing gap discovered within the Iranian context, the current study will examine productivity in food industry of Iran.

Methodology
The objective of this paper is identifying place of food industries in term of productivity compared to total industry, Therefore, at first index method will be applied to measure total productivity levels and then will utilize econometrics method for estimation of TFP growth. As said, production function expresses output as a function of the stock capital, employment, and a shift factor (t), time, where the latter proxies the effects of productivity and technical progress. The subscript t also represents time.
Assume that the argument "t" is separable from K and L; ( , ) This way, A t is referred to as exogenous, disembodied, and Hicks-neutral technical progress, and was measured by how output changes and time elapses with the input bundle held constant. Therefore, the notion of overall productivity can be reinterpreted as an index of all those factors other than labor and capital not explicitly accounted for but contributed to the generation of output.

Kendrick Index
Kendrick's index of total factor productivity for the case of value added as output, and two inputs can be written as: Where; A t is the value of index in a given year, T V is the added value; w and r denote the factor rewards of labor and capital respectively in the base year.

Parametric Approach
Parametric approach consists in econometric estimation of production functions to infer contributions of different factors and of an autonomous increase in production over time, independent of inputs. This later increase which is a shift over time in the production function, can be more properly identified as technological progress. It is one of the factors underlying productivity growth. Below commonly used specifications of production functions are given. Cobb-Douglas Specification: Where, Q, L, K and t refer to output, labor, capital and time. α  and β give factor shares respectively for labor and capital. A 0 describes initial conditions. Technological change takes place at a constant rate λ. It is assumed to be disembodied and Hicks-neutral, so that when there is a shift in the production function, K/L ratio remains unchanged at constant prices. Log-linear form this function can be written as:

Data sources
Annual data on output, value added, capital and labor for the food industries and total industries in two-digit were compiled for the period 1971-2006 from the Annual Survey of Manufacturing Industries published by the Statistical Centre of Iran. The variables were deflated by using price index of each group on the base year 1997 that published by Central Bank of Iran.

Empirical results
The levels of productivity between food industries and total industries are obtained by using Kendrick Index as exhibited on the equations below:   Table (1). Also trend of total productivity and labor productivity illustrated in figures (1&2).
The results of this study indicate that total factor productivity in food industries has been much lower than total industries (see figure 1), while labor productivity in food and total industries has been very close together, except that in recent years, total industries have been higher food industries (see figure 2).
To estimate of technological changes although panel data existed for food industries but due to lack of panel data for total industries, we have to use time series data for our estimations.
The Cob-Douglass production function form is applied for food industries as: And for total industries as: Where, food t Q is output of food industries and total t Q is output of total industries in the period 1971-2006.

R-squared=0.99 (1.8) (3.3) (4)
The results of the estimation show that Technological changes in food industries have been lower than total industries over the 1971-2006 periods. In general all results show that the productivity and Technological changes in food industries of Iran are not satisfactory and acceptable. This study was aimed to find out the reasons for this egregious difference between food and total industry of Iran in terms of labor productivity, total productivity and Technological changes. Several reasons have been found for this problem; one of the findings was that the capital per worker in food industry was lower than the total industry in 1995-2005 periods (see Table and Figure 3). The capital per worker has a positive relationship with labor and total productivity. The low capital per worker causes a decrease in productivity. Another reason for having low productivity in food industry was that the ratio of women workers to men workers in food industry was higher than total industry (See Table 3 and Figure 4). According to empirical evidences female employees' productivity is generally less than male employees' productivity (Verner, 2000), (Crepon, Deniau, & Perez-Duarte, 2002), (Kawaguchi, 2003) and (Liqin & Xiao, 2006). Finally, the differences between the educated workers in food industry and total industry can be one of the reasons for differential productivity in food and total industry (see Table 3 and Figure 5). Table 3 and Figure 4 show that the educated workers in food industries are less than in total industries in the overall period of 1995-2006. According to human capital theory, a higher education yields a higher productivity (Schultz, 1960). Empirical evidences related to human capital have proven this theory (Ballot, Fakhfakh, & Taymaz, 2001), (Stephan & Szalai, 2003), (Takii, 2003), (Lorraine, Reed, & Reenen, 2006) and (Biesebroeck, 2007).

Conclusion
In conclusion, the importance of food industries, i.e. Priority of the Non-oil Exports in Foreign Trade, Respond to Nutrition of Population and Prevention of Wastage, and lack of research related to productivity in food industry of Iran have led to this study on the productivity in food industry of Iran over the 1971-2006 periods. The examination of the levels of labor productivity, total productivity and technical changes in food industries and compared with the total industries of Iran showed that labor productivity and total factor productivity in food industries were lower than the average total industries over the period. Also the estimation of technical changes show that the measurement of technical change in food industries was 0.09 percent while for total industries was 0.16 percent over the period. There have been several reasons for this egregious difference between the food and total industry of Iran in terms of labor productivity, total productivity and Technological changes. The reasons were due to lower capital per worker, lower educated workers and higher women workers in food industry in respect to total industry.