ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Rural Wages during the 1990s: A Re-estimation

A major premise of economic reforms was the stimulus it was expected to give to the agriculture sector and to the demand for labour in rural areas. However, recent studies have given conflicting results regarding the trend in wages. In this paper, we re-estimate rural wage rates from the three NSS rounds for 1983, 1993-94 and 1999-2000 for 15 major states. Our results show that the growth rate of manual casual agricultural wages declined during the post-reform period with some differences at the state level. In the case of manual casual non-agricultural wages there was no decline in the growth rates at the all-India level. In the case of wages for all casual (manual and non-manual) wage labour, a decline was registered by both agricultural and non-agricultural wages in the 1990s. Analysis of the determinants of agricultural wages revealed that agricultural productivity, rural diversification, investment per hectare in agriculture, and percentage of agricultural labourers in the workforce were the key factors, with the impact of the latter two being lower in 1999-2000 compared to the earlier periods.

Rural Wages during the 1990s: A Re-estimation

A major premise of economic reforms was the stimulus it was expected to give to the agriculture sector and to the demand for labour in rural areas. However, recent studies have given conflicting results regarding the trend in wages. In this paper, we re-estimate rural wage rates from the three NSS rounds for 1983, 1993-94 and 1999-2000 for 15 major states. Our results show that the growth rate of manual casual agricultural wages declined during the post-reform period with some differences at the state level. In the case of manual casual non-agricultural wages there was no decline in the growth rates at the all-India level. In the case of wages for all casual (manual and non-manual) wage labour, a decline was registered by both agricultural and non-agricultural wages in the 1990s. Analysis of the determinants of agricultural wages revealed that agricultural productivity, rural diversification, investment per hectare in agriculture, and percentage of agricultural labourers in the workforce were the key factors, with the impact of the latter two being lower in 1999-2000 compared to the earlier periods.

RAVI SRIVASTAVA, RICHA SINGH

I Introduction

E
conomic reform and its impact on the well-being of the poor has been a widely debated topic in the recent past. Impact of the economic reforms on rural labour was largely expected to be indirect and entrenched in the stabilisation and structural measures. The liberalisation of agriculture and the lowering of tariffs and removal of trade controls were expected to provide a major fillip to agriculture, raising agricultural incomes and productivity through farm and non-farm diversification and also raising the demand for labour. At the same time, the thrust given to labour-intensive export-based growth was expected to increase the demand for non-farm labour. This was to have had a substantial beneficial impact on agricultural and rural wages, accelerating the reduction of poverty in rural India [Rao 1994].

Yet at the same time it was pointed out that stabilisation measures, which cut down on public investment, would have an unfavourable impact on the condition of rural labour. It would thwart public investment in rural infrastructure and since adequate rural infrastructure is a necessary precondition to realise the employment gains through the reform process the benefits would be limited. Further, lower investment in agriculture would reduce productivity in the sector and hence reduce demand for labour. Besides, with the shift in production away from coarse grains and cereals along with free play of market forces, prices of foodgrains would increase which would depress real wages.

Greater opportunities were expected to be generated in areas and regions where there was better infrastructure and diversification in rural areas and also within agriculture. Further, those endowed with better skills would have greater access to non-farm employment opportunities, which were to be generated after the reform process. Given the location, skill and capital bias of the reform measures, the gains from the reforms were likely to circumvent rural labour. This was expected to happen because of the characteristics of rural labour such as high incidence of poverty, assetlessness, poor access to health and educational facilities and deplorable living conditions. Besides lack of organisation and social security benefits for the group would further preclude them from making use of the opportunities provided by the structural reforms. Thus, with little skills and assets of their own and poor access to resources, rural labour households seem to have limited possibilities of sharing the gains of the market-oriented economic reforms. Further differences in infrastructure levels and pace of diversification would accentuate regional disparities in wage and employment levels. Gender disparities are also likely to be aggravated due to differences in education and skill levels. Recent studies [Radhakrishna and Ravi 2003; Deaton and Dreze 2002; Panchmukhi 2000] also provide empirical support to these fears.

Wage levels have come to be identified as an important indicator of the levels of living of rural labour and are also an important correlate of rural poverty. Some studies on trends in rural wages revealed that the tendency of wages to rise in the 1980s was reversed during the 1990s and the growth of wages registered a deceleration [Sen 1994; Parthasarthy 1996; Bhalla 1997; Unni 1997 and Sarmah 2001; Bhalla 2004; Srivastava and Singh 2005]. However, some recent studies do not support the trend of deceleration in real agricultural wages. Sharma (2001), for instance, using RLE data concluded that the agricultural wages did not witness a decline during the 1990s contrary to the findings of studies based on AWI data. Sundaram (2001a) also rejects the view that there has been a slowdown in the rate of growth of average daily wage earnings of adult labourer during 1990s. Thus, the debate regarding the impact of economic reforms on agricultural wages is still inconclusive. Use of different secondary data sources on rural wages has led to such conflicting trends.

Economic and Political Weekly September 23, 2006 Studies such as those by Himanshu (2005) have drawn attention to issues related to the credibility and comparability of these different sources.

In this paper, we seek to re-estimate rural wage rates in order to remove any discrepancies in the secondary data sources and thus reconcile the findings based on different sources. This would also ascertain if there has been any slowdown in the growth rate of rural wage rates in the post-reform period. We also seek to review the impact of the reforms on rural labour through analysing the performance of agricultural wages for the decade following the onset of structural adjustment and economic reforms. The paper is divided into five sections including this introductory section. Section II presents data and methodology followed by Section III that gives the trends in growth rates of real wage rates for agricultural and non-agricultural activities. Section IV discusses the trends in disparities in rural wages. Section V explains growth in real agricultural wages based on results of panel regression, which is followed by the concluding section.

II Data Sources and Methodology

Data on agricultural/rural wages are available from many sources. Out of these sources, the most frequently used source is the data given in Agricultural Wages in India (AWI), compiled and published by the directorate of economics and statistics, ministry of agriculture. AWI is the only source that provides time series data on agricultural wages. Rural Labour Enquiry (RLE) reports, published by the Labour Bureau are based on the quinquennial NSS surveys, and have also been used in a number of studies to analyse trends in both agricultural and non-agricultural wage earnings of rural labour households. The two sources differ not only in periodicity but also in methodology and coverage [Rao 1972]. Some recent studies [Sundaram 2001a and b; Bhalla 2004; Himanshu 2005] have used NSS employment and unemployment surveys to estimate agricultural wages because of the greater reliability, superiority in methodology and uniformity in concepts and definitions used in these surveys.

The aforementioned studies have given divergent results since they are based on different data sources, which differ in concepts and methodology. While studies which are based on AWI data such as those by Sarmah (2002) show a deceleration in agricultural wages in the 1990s, others such as those by Sundaram (2001a) that make use of NSS wage series report no indication of a slowdown in rural wages during the same period. Wage trends based on RLE such as in Srivastava and Singh (2005) show a very steep deceleration in the 1990s compared to the 1980s. This is in spite of the fact that NSS and RLE wage rates are drawn from same database, that is, the NSSO employment and unemployment surveys. There thus appears to be a great deal of incongruity in the trends in rural wages based on these sources during the last two decades.

Himanshu (2005) has undertaken an in-depth analysis of the reliability and comparability of the various secondary sources of data on rural wages, whereby he concludes that differences in concepts, methodology, choice of respondents and method of aggregation makes a direct comparison of wage trends based on these different sources inappropriate. Another significant finding of his analysis was the discrepancy found in the wage earnings published by RLE for the year 1983 (38th round). Recalculations of rural wage rates based on RLE specifications from unit record NSS data revealed that the wage estimates for 1983 published by RLE were underestimates, while for the 50th and 55th rounds, RLE estimates were exactly replicated. This greatly altered the picture, as the stupendous growth in rural wages in the 1980s appeared to be erroneous. The corrected estimates so derived were found to be in greater congruence with estimates based on NSS data derived by Sundaram (2001a).

The observations made by Himashu (2005) have made a reestimation of the rural wage rates a necessity in order to get a clear view of the trends in rural wages during the last two decades and to examine the deceleration reported in some studies. In this paper we make use of NSS employment and unemployment surveys to derive wage rates. We focus on agricultural wages for casual manual labour and compare the rate of growth of real wages in the pre- and post-economic reform period as. Wages have been computed for daily status casual labourers (code 41 and 51) in 15 to 59 years age group, engaged in manual work (as per the type of operation) in rural areas. Wages have been computed for both agriculture and non-agriculture operations. The computed wage rate is the average daily wage rate.

A close scrutiny of the wage data in the NSS data sets shows the existence of a large number of extreme values at both ends of the distribution. In order to remove any discrepancies arising due to presence of out of range values at both extremes in the data, we have resorted to trimming of the data set by excluding extreme upper-most and lower-most values. There are various ways in which the outliers could have been removed. One such way which is often attempted is to remove values which exceed mean +/- 3 standard deviation (SD). However, since the mean and SD are likely to be affected by the existence of the extreme values, we have tried trimming by excluding a similar proportion of cases at both extremes. Trimming has been tried at both 0.5 per cent and 1 per cent level. We therefore omit 0.5 or 1 per cent cases at the upper and lower end of the wage rates for all rounds at unweighted stage and for casual manual labourers.

It is evident from Table 1 that the 43rd round (1987-88) has the greatest discrepancy. The number of valid observations for that year is very small as the number of observations with zero wages is very high at 79 per cent. Also the share of the top 0.5 per cent and 1 per cent in the daily wage bill is very high. Other studies such as those by Himanshu (2005) have also cautioned against calculation of wage rates for this round.1 We have thus dropped the round for this paper.

Table 1: Summary of Data after Trimming of Data Sets

1983 1987-88 1993-94 1999-2000 (38th Rd) (43rd Rd) (50th Rd) (55th Rd)

Total no of observations 43306 13461 34802 50801 Missing 3 42 0 0 Zero wages 919 10682 17 22 Number of valid observations 42387 2779 34785 50779 Min +ve value (Rs) 0.0 0.2 0.0 2.0 Max +ve value (Rs) 1429.1 14322.3 3030.0 10080.0 Lower cut off (0.5 per cent) (Rs) 0.93 0.83 0.19 8.30 Lower cut off (1 per cent) (Rs) 1.40 1.39 0.33 9.70 Upper cut off (0.5 per cent) (Rs) 29.29 198.30 78.00 154.30 Upper cut off (1 per cent) (Rs) 23.75 114.30 68.57 139.30 Per cent share of top

0.5 per cent in daily wage bill 7 3 5 3 3

Per cent share of top 1 per cent in daily wage bill 9 3 7 4 5

In the 38th round, the lower end cut off point is slightly higher than the 43rd round and much higher than the 50th round, which suggests an upward bias in that round. While in the 50th round there appears to be a downward bias as the round has a lower cut off point than both 38th and 43rd round at the lower end while at the upper end too the cut off point is lower than the 43rd round, while in case of the 55th round the cut off point at both ends is much higher than the other three rounds. This has possibly pulled down the average wage rate for the 50th round, in which case the growth rates will be affected.

In the final analysis we have made use of the wage rate derived by trimming at the 1 per cent level. Trends have been analysed for 15 major states based on wage rates derived from the household-level data of the NSS employment and unemployment rounds for 1983, 1993-94 and 1999-2000. The nominal wage rates are converted into real wages by deflating with the Consumer Price Index for Agricultural Labour (CPIAL), published in the Labour Journal. Compound annual growth rates have been calculated for the period 1983 to 1999-2000, where 1983 to 1993-94 and 1993-94 to 1999-2000 are subperiods, which broadly correspond to pre- and post-reform periods.

We then use the state level data over the entire time period to see whether the determinants of wages have changed over time. In our regression analysis, we have pooled together our crosssection data across the three points of time and have used panel data analysis to obtain the results. By combining time series with cross-section data, panel data analysis gives more informative data, more variability, less collinearity among variables, more degrees of freedom and more efficiency. The technique of panel data analysis can take into account the heterogeneity between states by allowing for individual-specific variables and is also more suited to studying the dynamics of change. Thus, panel data analysis is more suited to better detect and measure effects that simply cannot be observed in pure cross-section or pure time series data. Our general functional form is as follows:

Ln(wit) = α +βXit + Eit where wit is real wage/daily earnings at 1999-2000 prices, X is the array of independent variables, β is the array of coefficients, i=1,2,3 …. 15 , for the 15 major states, and t = 1983, 1993-94 and 1999-2000.

III State Trends in Real Rural Wages

Growth rates for real wages for manual casual labour (15-59 years) in agricultural and non-agricultural operations are given in Tables 2 and 3 respectively.2 Trends in non-agricultural wages have also been discussed as it has been argued in some recent studies that an acceleration in growth rate in non-agricultural wages during the last decade has led to a shift of labourers to the sector and thus this shift is no longer a distress phenomenon as has been perceived in the past [Sundaram 2001b].

Trends in Wages in Manual Agricultural Operations

At the all-India level the annual growth rate for total real wages for manual casual labour in agriculture decelerated from

3.3 per cent per annum in 1983-94 to 2.7 per cent per annum in 1999-2000. Across the states, nine out of the 15 states registered lower growth rates in the post-reform period, with the sharpest decline being in West Bengal, Punjab (where growth rates were negative in the second period), Madhya Pradesh, Orissa and Assam. In Maharashtra the growth rate was a moderate 3.2 per cent per annum in both periods. A sharp rise in growth rates was registered in Kerala while Haryana, Bihar and Gujarat also showed moderately higher growth rates in the post-reform period.

Gender-wise the trends have been broadly similar across the states. Among the five states that showed higher growth rates in the second period, the growth rates for male casual labour have been higher in Kerala, Andhra Pradesh and Gujarat compared to female casual labour. While among the states that showed a decline in growth rates in the second period the decline has been perceptibly sharper for female casual labour in Punjab, Rajasthan Tamil Nadu and Karnataka. Growth rate at the all-India level, for females has been marginally higher than that of males in both periods

During 1983-94, several states with low initial wages, such as Karnataka, Maharashtra, Madhya Pradesh, Orissa and Tamil Nadu experienced above an average growth rate in real wages. For all states taken together, there was a negative and significant correlation (-0.65) between initial wage level and the growth of real wages during 1983-93. However, during 1993-2000,

Table 2: State-wise Growth Rate of Real (at 1999-2000 Prices) Rural Wages in Manual Agricultural Operationsfor Casual Labourers (15-59 Years) – 1983 to 1999-2000

(in per cent a year)

States Male Female Total 1983-1994 1994-2000 1983-2000 1983-1994 1994-2000 1983-2000 1983-1994 1994-2000 1983-2000

AP 2.7 4.0 3.2 3.2 3.2 3.2 3.0 3.6 3.2 Assam 2.3 0.5 1.7 2.1 0.2 1.4 2.2 0.4 1.6 Bihar 3.0 4.4 3.5 3.5 5.2 4.1 3.2 4.6 3.7 Gujarat 1.1 3.5 2.0 1.9 1.8 1.9 1.4 2.9 2.0 Haryana 1.0 1.5 1.2 -0.8 2.6 0.4 0.4 2.0 1.0 Karnataka 4.1 3.6 3.9 4.2 2.8 3.7 4.1 3.3 3.8 Kerala 1.4 6.3 3.2 2.0 4.8 3.0 1.7 5.9 3.2 MP 4.1 0.6 2.8 3.8 1.2 2.8 4.1 0.7 2.8 Maharashtra 3.1 2.9 3.0 3.7 3.5 3.6 3.3 3.2 3.3 Orissa 4.2 1.1 3.0 4.0 2.3 3.3 4.1 1.2 3.1 Punjab 2.5 -0.3 1.5 3.9 -1.1 2.0 2.6 -0.3 1.6 Rajasthan 2.5 2.6 2.5 4.4 1.3 3.3 3.2 2.4 2.9 TN 5.6 4.2 5.1 5.7 3.9 5.1 5.5 4.1 5.0 UP 3.0 2.4 2.8 2.6 2.4 2.5 3.0 2.1 2.7 WB 4.0 -2.1 1.8 3.9 -2.1 1.6 4.0 -2.1 1.7 India 3.3 2.7 3.0 3.4 2.9 3.2 3.3 2.7 3.1

Source: Computed for wage rates derived from the NSS Employment/Unemployment Rounds 1983, 1993-94 and 1999-2000.

Economic and Political Weekly September 23, 2006 initial wage levels (in 1993-94) and growth of real wages during 1993-94 to 1999-2000 showed a much lower negative correlation (-0.22). Thus during the pre-reform period, low wage rate states experienced a significantly faster growth in agricultural wages compared to the high wage rate states, this trend, however, weakened considerably in the post-reform period.

Trends in Wages in Manual Non-Agricultural Operations

Non-agricultural wages for manual work in rural areas grew at a rate of 4 per cent during 1983-94 and 4.1 per cent during 1994-99 at the all-India level. Though the growth rate has been reasonably robust at the all-India level, across the states there are some variations. Seven out of the 15 states register a lower growth rate in the post-reform period. The decline is steepest in West Bengal followed by Assam, Bihar and Orissa. The rest of the states show a marginally lower growth rate in the second period. Among the eight states showing a rise in the second period states of Andhra Pradesh and Tamil Nadu show a sharp rise. In Madhya Pradesh growth rates are the same in both periods while for Punjab and Rajasthan a very small rise is experienced in the growth rates for the post-reform period. The performance of the southern states has been above average in both periods, while low wage rate states of Bihar, Orissa and West Bengal perform better in the first period.

Gender-wise, the trends differ for non-agricultural wages unlike the case of agricultural wages. Not only has the growth rate for female casual labour in non-agriculture been higher than for the males in both period, but also there has been acceleration in case of the former compared to a small decline in case of the latter. Nine states show a lower growth rate in the second period for male casual labour while the number for female casual labour is seven. Growth rate of wages for female casual labour in nonagriculture have been exceptionally high in Haryana and Punjab, while Maharashtra, Kerala, Madhya Pradesh and Andhra Pradesh also record high growth in the post-reform period. In the states of Bihar and Uttar Pradesh, growth rate for female casual labour in non-agriculture has been higher than their male counterparts in both periods.

As in the case of agricultural manual wages, non-agricultural wages for low wage rate states such as Tamil Nadu, Maharashtra, Karnataka and Bihar record above average growth rate in real wages in the pre-reform period. During this period there was a negative and significant correlation (-0.64) between initial wage level and the growth of real wages. However, during 1993-2000 initial wage levels (in 1993-94) and growth of real wages during

Table 3: State-wise Growth Rate of Real (at 1999-2000 Prices) Rural Wages in Manual Non-Agricultural Operations forCasual Labourers (15-59 Years) – 1983 to 1999-2000

(in per cent a year)

States Male Female Total 1983-1994 1994-2000 1983-2000 1983-1994 1994-2000 1983-2000 1983-1994 1994-2000 1983-2000

AP 2.7 5.4 3.7 1.9 3.7 2.5 2.5 6.4 3.9 Assam 2.9 -0.1 1.8 3.1 0.6 2.2 2.8 0.5 2.0 Bihar 4.2 2.2 3.5 5.5 3.4 4.7 4.6 2.8 3.9 Gujarat 1.6 1.5 1.6 1.4 2.9 1.9 1.9 1.2 1.6 Haryana 2.3 2.6 2.4 0.2 9.0 3.3 2.2 2.7 2.4 Karnataka 3.5 5.2 4.1 5.5 5.0 5.3 4.3 5.9 4.8 Kerala 3.4 4.2 3.7 3.6 6.1 4.5 3.9 4.9 4.3 MP 2.5 1.8 2.2 2.1 4.0 2.8 2.4 2.4 2.4 Maharashtra 4.4 3.7 4.2 4.1 7.3 5.3 5.0 4.7 4.9 Orissa 4.3 3.3 3.9 3.7 1.3 2.8 4.3 3.1 3.9 Punjab 2.2 1.2 1.8 0.9 8.6 3.6 1.8 2.0 1.9 Rajasthan 2.6 3.9 3.1 3.3 3.7 3.5 3.1 3.9 3.4 TN 4.8 7.4 5.8 6.1 5.4 5.9 5.5 7.6 6.3 UP 2.4 1.8 2.2 4.6 4.3 4.5 2.5 2.0 2.3 WB 4.4 -2.3 1.9 6.1 -2.6 2.8 4.5 -1.7 2.2 India 3.7 3.6 3.7 4.2 5.1 4.5 4.0 4.1 4.1

Source: Computed for wage rates derived from the NSS Employment/Unemployment Rounds 1983, 1993-94 and 1999-2000.

Table 4: State-wise Growth Rate of Real (at 1999-2000 Prices) Rural Wages in Agricultural Operations for Casual Labourers(15-59 Years) – 1983 to 1999-2000

(in per cent a year)

States Male Female Total 1983-1994 1994-2000 1983-2000 1983-1994 1994-2000 1983-2000 1983-1994 1994-2000 1983-2000

AP 2.8 4.1 3.3 3.2 3.3 3.2 3.0 3.7 3.3 Assam 2.3 1.7 2.1 2.1 0.5 1.5 2.3 1.5 2.0 Bihar 3.1 4.3 3.5 3.5 5.2 4.1 3.2 4.6 3.7 Gujarat 1.3 3.9 2.2 1.7 2.0 1.8 1.4 3.3 2.1 Haryana 1.0 2.0 1.4 -0.9 2.5 0.4 0.5 2.5 1.2 Karnataka 4.2 3.6 4.0 4.2 2.8 3.7 4.2 3.4 3.9 Kerala 1.3 5.8 2.9 1.7 4.5 2.7 1.5 5.6 3.0 MP 4.1 1.0 3.0 3.8 1.4 2.9 4.1 1.1 3.0 Maharashtra 3.3 3.1 3.2 3.7 3.5 3.6 3.4 3.4 3.4 Orissa 4.3 1.2 3.2 4.0 2.4 3.4 4.2 1.4 3.2 Punjab 2.4 0.1 1.6 3.8 -3.3 1.2 2.6 -0.1 1.6 Rajasthan 2.6 3.3 2.8 4.6 1.3 3.4 3.3 3.0 3.2 TN 5.5 4.4 5.1 5.8 4.0 5.1 5.5 4.4 5.1 UP 3.2 3.2 3.2 2.6 2.4 2.5 3.2 2.8 3.0 WB 4.1 -2.1 1.8 3.6 -1.8 1.6 4.0 -2.0 1.8 India 3.4 3.0 3.2 3.4 3.0 3.2 3.4 3.0 3.3

Source: Computed for wage rates derived from the NSS Employment/Unemployment Rounds 1983, 1993-94 and 1999-2000.

1993-94 to 1999-2000 showed a lower negative correlation (-0.33). Thus like agricultural wages, non-agricultural wages for low wage rate states experienced a faster growth compared to the high wage rate states during the pre-reform period. The post-reform period saw a weakening of this trend.

Thus, manual casual agricultural wages show a deceleration in the post-reform period, which is shared by both male and female manual agricultural labour. But in case of manual casual non-agricultural wages, there is no deceleration at the all-India level though at the state level seven states show a lower growth rate in the second period. For non-agricultural wages, male manual wage labour shows a marginally lower growth rate in the 1990s while for female manual labour in non-agriculture, the growth rate has been higher than male labour in both periods and has also accelerated in the post-reform period. The wage gap between the male and female manual non-agricultural labour thus appears to have narrowed down in the last two decades.3

Growth Rate of Casual Wages

Though the focus of this paper is on trends in manual casual wages, we have also estimated the growth rate of all casual labour (manual as well as non-manual labour). This is in order to determine whether the composition of the casual labour force between its manual and non-manual components, has had any influence on the growth rate of wages.

Trends in Wages for Casual Workers in Agricultural Operations

Table 4 presents the results for growth rates of wages of casual rural workers in agricultural operations. The trends in casual rural wages in agriculture are broadly similar to those in case of wages for casual manual work in agriculture. At the all-India level, annual growth rate for total real wages for casual labour in agriculture shows a slowdown from 3.4 per cent per annum 198394 to 3 per cent per annum in 1983-94. The corresponding slowdown, it would be recalled, was steeper in case of manual agricultural wages (from 3.3 per cent to 2.7 per cent).

At the all-India level both males and females register identical growth rates for the two periods. Across the states also the direction of trends is broadly similar. Nine out of the 15 states registered lower growth rates in the post-reform period. West Bengal, as in case of manual casual agricultural wages registered the sharpest decline and a negative growth rate in the second period. Punjab is the other state, which shows a negative growth rate in real casual wages during 1993-94 to 1999-2000. Among the six states showing higher growth rates in the second period Kerala registered the steepest rise while Haryana, Bihar and Gujarat also showed moderately higher growth rates in the post-reform period. Eight states in case of males and 10 in case of females register a lower growth in casual agricultural wages in the 1990s.

Trends in Wages for Casual Workers in Non-Agricultural Operations

At the all-India level, wages for casual workers in non-agricultural operations showed a marginally lower growth rate in the post-reform (Table 5), coming down from 4 per cent per annum in 1983-94 to 3.9 per cent per annum in 1994-99. This can be compared to manual non-agricultural wages, which showed a slight rise in growth rate in the post-reform period, from 4 per cent in the first period to 4.1 per cent in the second.

Across the major states, six states registered lower growth rates for casual non-agricultural wages in the second period, while one (Karnataka) registered similar growth rate in wages in both periods. West Bengal registered a negative growth rate in the second period while Bihar, Orissa, Assam, Maharashtra and Uttar Pradesh registered lower growth rates in the second period. Among the states showing a higher growth rate in the second period, Andhra Pradesh recorded the highest acceleration followed by Tamil Nadu, Kerala and Gujarat.

As observed in case of casual manual non-agricultural wages, wages for female casual workers in non-agricultural operations also show a significantly higher growth rate at the all-India level, during 1994-2000 . On the other hand, the growth rate of male casual wages shows a steeper decline in the 1994-2000 period compared to casual manual wages (from 3.7 per cent during 1983 to 1993-94 to 3.4 per cent during 1993-94 to 1999-2000). Among the 15 major states, seven states show an increase in female wage growth rates in the second period and a similar number show an increase in male wage growth rates. Growth rates for both male and female wages declined in the second period in Assam, Bihar, Orissa, UP and West Bengal and increased in AP, Gujarat, Haryana and Karnataka.

Thus, the trends in wages for casual workers in both agriculture

Table 5: State-wise Growth Rate of Real (at 1999-2000 Prices) Rural Wages in Non-Agricultural Operations for CasualLabourers (15-59 Years) – 1983 to 1999-2000

(in per cent a year)

States Male Female Total 1983-1994 1994-2000 1983-2000 1983-1994 1994-2000 1983-2000 1983-1994 1994-2000 1983-2000

AP 3.0 5.0 3.7 2.3 4.0 2.9 2.8 5.6 3.8 Assam 2.5 1.0 2.0 3.5 1.4 2.8 2.6 1.4 2.1 Bihar 4.3 1.7 3.4 5.3 3.4 4.6 4.6 2.3 3.8 Gujarat 1.4 2.8 1.9 -0.1 4.2 1.5 1.4 2.7 1.9 Haryana 2.3 2.5 2.4 -0.6 7.2 2.2 2.2 2.5 2.3 Karnataka 3.7 4.0 3.8 5.6 4.3 5.1 4.6 4.6 4.6 Kerala 3.0 4.2 3.4 3.2 6.1 4.2 3.4 4.9 4.0 MP 2.7 2.0 2.4 2.1 4.2 2.9 2.5 2.6 2.5 Maharashtra 4.6 3.1 4.0 4.0 6.4 4.8 5.0 4.3 4.8 Orissa 4.4 2.7 3.8 3.7 2.4 3.2 4.3 2.8 3.8 Punjab 1.9 1.8 1.9 0.6 5.3 2.3 1.7 1.9 1.8 Rajasthan 2.5 4.1 3.1 3.9 3.2 3.7 3.1 4.1 3.5 Tamil Nadu 4.8 6.8 5.5 6.4 5.5 6.1 5.4 7.0 6.0 UP 2.5 2.2 2.4 4.2 2.9 3.8 2.6 2.3 2.5 WB 4.4 -2.1 2.0 5.8 -2.3 2.8 4.3 -1.4 2.2 India 3.7 3.4 3.6 4.1 4.8 4.3 4.0 3.9 4.0

Source: Computed for wage rates derived from the NSS Employment/Unemployment Rounds 1983, 1993-94 and 1999-2000.

Economic and Political Weekly September 23, 2006 and non-agricultural operations are broadly similar to the trends observed in case of wages for manual casual workers in rural areas. However, in the case of all casual workers, there is a more pronounced decline in the second period in the case of male nonagricultural wages resulting in a small overall decline during 1993-94 to 1999-2000 in these wages. Agricultural wages show a decline for both casual manual and total casual wages but the decline is sharper at the all-India level for the former. Genderwise, as noted above, for both manual as well as total casual wages, males and females show a slowdown in the second period for growth rate in agricultural wages, but in case of nonagricultural wages, female wages are higher in the second period, while male wages are lower. The state level pattern is considerably varied, as shown by the preceding analysis.

Trends in Agricultural Wages: A Comparison of Results Based on Different Sources of Wage Data

Table 6 gives the growth rates for agricultural wages for male casual labour based on different sources of wage data. The wage rates from the different sources are not strictly based on same concepts and methodology, yet the wage trends can be compared to get an overall picture. Overall the trends are similar except that the RLE based trends show a very sharp deceleration in the 1990s when compared to trends based on AWI4 and NSS wage rates. RLE results for 1983, as indicated by Himanshu (2005), are erroneous and give underestimates for the year, which propels such high growth rates for the 1980s.

Results based on NSS are better correlated with the results based on AWI.5 A deceleration is registered at the all-India level by all three sources. The deceleration is sharpest based on RLE (from 7.2 per cent to 2.9 per cent) followed by AWI (3.7 per cent to 2.8 per cent) and lowest in case of NSS (3.3 per cent to 2.7 per cent). State-wise there are some differences. RLE shows a decline for all states except Bihar. Eight states show a deceleration in growth rate of agricultural wages in the post-reform period, based on all the three sources. Sharpest decline is in West Bengal, followed by Punjab, Madhya Pradesh and Orissa. The states of Kerala, Gujarat and Haryana show acceleration in growth rates based on NSS and AWI while Bihar shows a rise in case of AWI and NSS.

A look at the trends based on the three different sources confirms spatially widespread deceleration in agricultural wages in the post-reform period. High wage rate states of Kerala and Haryana and typically non-agrarian states of Gujarat and Rajasthan register acceleration in agricultural wage rates during the period.

Comparison of Findings Based on NSS Data

Estimates of rural wages made by Sundaram (2001a) and later by Himanshu (2005) do not reflect a deceleration in rural wages in the 1990s. Table 7 gives a comparison of these estimates. Himanshu’s estimates are for rural labour households while Sundaram’s wages are for casual labour in age group 15-59. Both these studies did not resort to trimming of unit record data sets and thus the results include discrepancies which may arise due to presence of out of range values and incorrect data entries. The underestimation of wages for 1993-94 thus has not been corrected for and probably this is the reason why these estimates do not bring out the deceleration in rural wages in the 1990s. Our estimates for 1983 are lower than those estimated by Sundaram (2001a) and Himanshu (2005), the gap being greater in the case of non-agricultural wages. As discussed earlier, the 1983 wages had an upward bias. The share of the top 1 per cent in the daily wage bill was much higher at 9 per cent compared to 4 per cent and 5 per cent in the case of 1993-94 and 1999-2000 respectively. Thus, by adjusting for this upward bias the average wage rate has been pulled down. For 1993-94, the wages are fairly close in the case of agricultural wages while in 1999-2000 the difference is higher in non-agricultural wages with our estimates being lower. Thus the adjustment in the upward and downward bias has corrected the average wage rate and the growth rates.

It would also be noted that the growth rates discussed above are based on point-to-point estimates, since NSS results are available only at discrete time intervals. These growth rates could be different from estimates of trend growth rates, which are available from monthly or annual data. For agricultural wages, wage data is available at monthly intervals from the AWI. In

Table 6: Growth Rate of Male Agricultural Wages from Different Sources

(Per cent a year)

States 1983 to 1993-94 to 1983 to 1993-94 1999-2000 1999-2000 NSS RLE AWI* NSS RLE AWI NSS RLE AWI*

Andhra

Pradesh 2.7 6.2 3.2 4.0 4.3 0.5 3.2 5.5 2.3 Assam 2.3 4.2 2.4 0.5 0.7 0.5 1.7 2.9 1.8 Bihar 3.0 1.1 3.8 4.4 13.1 1.7 3.5 5.3 3.1 Gujarat 1.1 4.9 1.4 3.5 3.4 9.3 2.0 4.3 4.3 Haryana 1.0 6.8 2.5 1.5 4.4 3.5 1.2 5.9 2.9 Karnataka 4.1 8.5 2.8 3.6 4.0 3.4 3.9 6.8 3.1 Kerala 1.4 6.1 2.5 6.3 5.9 7.4 3.2 6.0 4.4 Madhya

Pradesh 4.1 7.7 5.0 0.6 1.5 1.3 2.8 5.4 3.8 Maharashtra 3.1 7.8 5.8 2.9 2.5 -0.5 3.0 5.8 3.7 Orissa 4.2 8.9 6.3 1.1 0.9 -1.4 3.0 5.9 3.7 Punjab 2.5 5.7 3.4 -0.3 -0.2 -1.7 1.5 3.6 1.6 Rajasthan 2.5 7.5 -0.4 2.6 3.2 5.8 2.5 5.9 1.8 Tamil Nadu 5.6 9.3 5.2 4.2 4.4 5.4 5.1 7.5 5.5 Uttar Pradesh 3.0 8.0 3.9 2.4 2.3 2.0 2.8 5.9 3.3 West Bengal 4.0 8.7 7.4 -2.1 -2.7 -2.1 1.8 4.4 4.1 India 3.3 7.2 3.7 2.7 2.9 2.8 3.0 5.5 3.5

Note: Wage rate for AWI is for the year 1982-83.

Source: NSS wage rate has been computed from the EUE survey 1983/19992000, Rural Labour Enquiry (RLE) Reports and Agricultural Wages in India (AWI).

Table 7: Male Real Rural Wages: A Comparison of Results

(at 1999-2000 prices)

Wage Level (Rs) Growth Rate (Per Cent a Year) 1983 1993-94 1999-2000 1983-94 1993-99

Casual labour in agriculture

Our 24.6 34.9 41.6 3.4 3.0 Himanshu* 25.7 34.3 40.5 2.8 2.8 Sundaram 25.8 34.3 40.4 2.8 2.8

Casual labour in non-agriculture

Our 30.6 44.9 54.9 3.7 3.3 Sundaram 37.3 47.8 59.5 2.4 3.7

Manual labour in agriculture

Our 24.4 34.2 40.2 3.3 2.7 Himanshu** 25.6 33.9 40.2 2.7 2.9 Sundaram 25.8 34.3 40.4 2.7 2.8

Manual labour in non-agriculture

Our 30.6 44.9 55.4 3.7 3.6 Sundaram 37.4 NA 60.3 NA NA

Notes: Wages for Sundaram have been converted to 1999-2000 prices.

* The figures are for all age groups. ** The figures are for ALHH.

Figure 1: Trends in Rural Real Wage Index ofUnskilled Labour in UP

190 170 150 130 110 90 70

2003-04 2002-03 2001-02 2000-01 1999-00 1998-99 1997-98 1996-97 1995-96 1994-95 1993-94 1992-93 1991-92 1990-91 1989-90 1988-89 1987-88 1986-87 1985-86 1984-85 1983-84 1982-83 1981-82

Srivastava and Singh (2005) we had obtained trend growth rates for the pre-reform and post-reform periods from annualised and averaged state-level data to show that the annual growth rate of real male agricultural wages declined from 4.9 per cent during 1981-91 to 2.6 per cent during 1992-2002.

For non-agricultural wages, no similar series is available at the all-India level. However, information on unskilled non-agricultural wages is collected on a regular basis in a few states. For example, the directorate of economics and statistics, Uttar Pradesh provides quarterly data on unskilled rural and urban labour.

The figure gives the trend in real wage index of male unskilled labour in Uttar Pradesh from 1981-82 onwards, based on this data. It is evident that the wages peaked around 1991 and thereafter a downward trend set in, which continues till 1999. Trend growth in unskilled male wages, using a modified kinked exponential function, shows a decline from 2.3 per cent per year during 1981-91 to 0.03 per cent per year during 1992-2004. The point-wise growth rate in unskilled wages was 2.3 per cent during 1983 to 1993-94 and 0.5 per cent during 1993-94 to 1999-2000. These figures show a sharper decline in growth rate in Uttar Pradesh compared to the NSS estimates presented in Table 2.

The issue at stake here is that point-to-point growth rates may reflect the impact of idiosyncratic factors on wages in any given year, which is not the case when trend growth rates are estimated.

IV Trends in Disparities

Interstate Variation in Rural Wages

Trends in interstate variation in real wages have been examined by calculating the coefficient of variation (CV) for the three points of time (Table 8). Interstate variation in agricultural wages appear to be much higher than in the case of non-agricultural wages. The coefficient of variation in real wages in agricultural operations showed a substantial decline in the pre-reform period followed by a rise in the post-reform period. Across gender the trends differ. For males, the CV declined from 38 per cent in 1983 to 30 per cent in 1993-94, but once again rose to 34 per cent in 1999-2000. The variation was observed to be higher for females; however there has been a continuous decline in the CV in both periods. Though there was a rise in CV in the post-reform period, it was still lower in 1999-2000 than what it was in 1983.

In the case of non-agricultural earnings the trend is similar as for agricultural earnings with variations first declining and then rising again in 1999-2000. The CV for males after rising initially declined to the same level in 1999-2000 as it was in 1983. In the case of females, the trend is similar with the CV coming down from 30 per cent in 1983 to 19 per cent in 1993 and then rising again to 21 per cent in 1999-2000. The decline was sharper in case of females compared to male casual labour.

Considering the period as a whole there has been a decline in inter-state variation in real earnings of rural wage earners. However, the reform period seems to have witnessed a rise in CV.

Gender Disparities in Rural Wages

Gender disparity in earnings has been analysed on the basis of the ratio of female wage levels to male wage levels. Table 9 gives the ratio of female/male rural wages for the period 1983 to 1999-2000.

At the all-India level the female-male wage ratio for agricultural wages showed lower disparity compared to non-agricultural wages. The ratio at the all-India level, showed only a marginal change in the case of agricultural wages in both periods. In the case of non-agricultural wages there was a consistent rise between 1983 (59.7 per cent) and 1999-2000 (68.1 per cent), indicating a narrowing of gender disparities. The rise was sharper in the post-reform period. This trend is reflective of the findings of trends in the growth rate whereby the gap between the female and male non-agricultural wage has narrowed down in the recent past.

Table 9: Ratio of Female and Male Rural Wages,1983 to 1999-2000

St ates Agriculture Non-Agriculture
1983 1993-94 1999 1983 1993-94 1999
2000 2000
AP 66.8 70.3 67.1 71.0 65.1 59.0
Assam 89.6 87.4 85.7 62.1 62.9 65.7
Bihar 82.0 86.2 90.6 67.7 77.6 83.1
Gujarat 89.4 96.5 87.5 67.5 66.0 71.6
Haryana Karnataka 100.3 70.7 83.2 71.2 88.9 67.9 72.3 51.0 58.2 62.7 83.4 62.0
Kerala 69.8 73.9 67.8 51.3 52.3 58.3
MP 84.8 82.3 85.6 72.5 69.9 79.6
Maharashtra 59.6 63.5 65.7 60.8 59.1 72.3
Orissa 74.5 72.9 78.3 82.7 78.4 69.7
Punjab Rajasthan 78.0 67.3 89.9 81.3 85.6 75.4 56.2 67.1 49.2 72.4 75.2 71.7
TN 57.6 58.4 57.5 49.0 55.8 49.7
U P 79.4 75.7 75.6 60.9 76.8 88.6
WB 86.5 84.9 84.8 57.8 68.1 66.9
India 69.9 70.4 71.3 59.7 62.4 68.1

Source: Computed for wage rates derived from the NSS Employment/ Unemployment Rounds 1983, 1993-94 and 1999-2000.

Table 8: Coefficient of Variation (Per Cent) for Rural Wages 1983 to 1999-2000

Male Female Total
1983 1993-94 1999-2000 1983 1993-94 1999-2000 1983 1993-94 1999-2000
Agriculture 37.6 29.9 33.5 44.3 35.1 33.6 40.9 32.7 35.0
Non-agriculture 24.9 21.4 23.0 29.9 18.9 21.3 27.1 20.8 21.9
Total 34.4 28.8 31.4 41.7 31.4 30.4 37.7 31.2 32.9
Source: Computed for wage rates derived from the NSS Employment/Unemployment Rounds 1983, 1993-94 and 1999-2000.
Economic and Political Weekly September 23, 2006 4059

The southern states of Andhra Pradesh, Karnataka, Kerala, Maharashtra and Tamil Nadu have a very low ratio indicating a higher gender disparity in rural wages in these states. In the case of non-agricultural wages, six states showed a decline in the ratio in the pre-reform period. The largest fall was in case of Haryana, with a much lower decline in case of Assam, Madhya Pradesh, Uttar Pradesh, West Bengal and Orissa, while the highest improvement in ratio was recorded in Rajasthan, Punjab and Gujarat. In the post-reform period 11 states show a decline in the wage ratio, with Kerala showing the highest fall. Overall, Bihar, Maharashtra, Punjab and Rajasthan have shown a major improvement in the ratio, indicating a decline in gender disparity in agricultural wages, while Haryana, Karnataka, Assam and Uttar Pradesh witnessed a worsening of gender disparities.

In the case of non-agricultural wages for the entire period, there has been a visible decline in gender disparity at the national level. However, some differences in the pattern of change are observed among states. In the pre-reform period eight states show a decline in the ratio with a major decline in case of Haryana, Andhra Pradesh and Orissa. A major improvement in the ratio was recorded in Uttar Pradesh, West Bengal, Karnataka and Tamil Nadu during the post-reform period, six states show a worsening of gender disparities. Haryana, Punjab, Maharashtra, Uttar Pradesh and Madhya Pradesh saw a decline in gender disparities, while Orissa, Tamil Nadu and Andhra Pradesh showed a rise. Punjab, Uttar Pradesh, Haryana, Bihar, Karnataka, West Bengal and Maharashtra witnessed a continuous decline in gender disparities throughout the period.

Overall, Bihar, Maharashtra, Punjab, Rajasthan and Madhya Pradesh recorded a decline in gender disparities in rural wages

Table 10: Ratio of Non-Agricultural and Agricultural Wages 1983 to 1999-2000

States Male Female Total 1983 1993-1999-1983 1993-1999-1983 1993-199994 2000 94 2000 94 2000

AP 1.2 1.2 1.3 1.2 1.1 1.1 1.3 1.2 1.4 Assam 1.2 1.2 1.2 0.8 0.9 0.9 1.1 1.2 1.2 Bihar 1.3 1.4 1.3 1.0 1.3 1.2 1.2 1.4 1.3 Gujarat 1.4 1.5 1.3 1.1 1.0 1.1 1.4 1.4 1.3 Haryana 1.0 1.1 1.2 0.7 0.8 1.1 1.0 1.2 1.2 Karnataka 1.4 1.3 1.5 1.0 1.2 1.3 1.4 1.4 1.6 Kerala 0.9 1.1 1.0 0.7 0.8 0.8 0.8 1.0 1.0 MP 1.5 1.3 1.4 1.3 1.1 1.3 1.5 1.3 1.4 Maharashtra1.2 1.4 1.4 1.2 1.3 1.6 1.3 1.5 1.6 Orissa 1.2 1.2 1.3 1.3 1.3 1.2 1.2 1.2 1.4 Punjab 1.0 1.0 1.1 0.7 0.5 0.9 1.0 0.9 1.1 Rajasthan 1.0 1.0 1.1 1.0 0.9 1.0 1.1 1.0 1.1 TN 1.2 1.1 1.3 1.0 1.0 1.1 1.2 1.2 1.5 UP 1.4 1.3 1.2 1.0 1.3 1.4 1.4 1.3 1.3 WB 1.1 1.1 1.1 0.7 0.9 0.9 1.0 1.0 1.1 India 1.3 1.3 1.4 1.1 1.2 1.3 1.3 1.4 1.5

Source: Computed for wage rates derived from the NSS Employment/ Unemployment Rounds 1983, 1993-94 and 1999-2000.

Table 11: Determinants of Real Agricultural Wages

Dependent: Log of Wage Rate in Manual Agricultural Operations

Coef Stand t-Value p-Value Coef

Ln NSDPAG per ag worker 0.33 0.46 6.6 0.00 Ln total expenditure in agriculture

per ha 0.10 0.40 7.2 0.00 Rural non-farm workers (per cent) 0.012 0.27 4.1 0.00 Agricultural labour (per cent) -0.012 -0.36 -6.2 0.00 Constant -0.146 . -0.3 0.76

(p>F=0.000) R2=0.91 AR2=0.90 Df=42

during the last two decades. The southern states of Kerala, Tamil Nadu and Andhra Pradesh continued to show little improvement in gender equality.

Differentials in Earnings in Agricultural and Non-Agricultural Operations

Table 10 shows the ratio of non-agricultural wages to agricultural wages for manual casual labour for the period 1983 to 1999-2000.

Overall, non-agricultural wages are found to be higher than agricultural wages, except in a few cases. In Haryana, Kerala, Punjab, Rajasthan and West Bengal, agricultural wages and nonagricultural wages have been at the same level. The ratio was as low as 0.5 in the case of females in 1993-94. In the states of Assam, Haryana, Kerala, Punjab and West Bengal agricultural wages have been generally higher than non-agricultural wages particularly for females.

On average non-agricultural earnings are 1.3 times the agricultural wages. Furthermore, the whole period shows a rising tendency in wage disparities. At the all-India level the ratio has risen from 1.3 in 1983 to 1.5 in 1999-2000, which was 1.3 in 1983 to 1.4 in 1999-2000 for males, while for females it has risen from 1.1 to 1.3 during the same period.

V Determinants of Agricultural Wages

Panel regression was used to ascertain determinants of manual agricultural wages within the demand/supply framework, used in a number of studies [Deepak Lal 1976; Bardhan 1984; Bhalla 1993; Parthasarthy 1996; Haque 1998; Sharma 2001; Sarmah 2002; Srivastava and Singh 2005]. Regression has been attempted with several explanatory variables that affect the demand and supply of manual agricultural labour particularly in the context of the economic reforms initiated in 1991. These broadly represent agricultural productivity, rural diversification, investment in agriculture, and supply of labour. The following variables have been used in the regression model:

  • (1) Agricultural Productivity: The Net State Domestic Product (in ‘000 Rs) in agriculture per agricultural worker (NSDPAGW);
  • (2) Non-farm Diversification: Percentage share of rural nonfarm workers in total rural workforce (RNFWK);
  • (3) Investment: Total expenditure per ha in agriculture (in Rs 00) (TEXPHA),6 and
  • (4) Labour Supply: Percentage of agricultural labourers in the rural workforce (PAGLAB).
  • In our regression analysis, we have pooled together our crosssection data across three points of time and 15 major states. Regression was tried with both the log-lin model and log-log model. The model with the time effect and time interaction in both cases did not yield significant results for the dummy variables and the interaction variables. The adjusted R2 also did not show an improvement.

    The results reported here pertain to the log-log form of the model. In this model we have taken log of NSDPAGWK and TEXPHA as the other two variables are already in percentages. The results can thus be interpreted in percentage terms. The results are given in Table 11. The adjusted R2 in this case is 0.9, reflecting the high explanatory power of the model. All variables are highly significant. The coefficients show that (a) an increase of agricultural income per worker by 1 per cent results in an increase of 0.33 per cent in real manual agricultural wage; (b) a one per cent increase in total expenditure in agriculture results in an increase of 0.10 per cent in real wages; (c) a 1 per cent point increase in the percentage of non-farm workers in the rural workforce results in a 1.2 per cent increase in real agricultural wages; and finally (d) an increase of 1 per cent point in the percentage of agricultural labourers to rural workforce results in a decline of 1.2 per cent in real wages.

    VI Conclusion

    The main aim of this paper has been to re-estimate rural wages in India from the three NSS rounds for 1993, 1993-94 and 19992000 and to compare the estimated growth in real wages with existing results. Rural wages are computed and published from time to time in the RLE reports of the ministry of labour. These estimates, which are for rural labour households, are also based on the NSS rounds. However, separate estimates based on these rounds have raised doubts regarding the RLE wage estimates, particularly for 1983. We have therefore relied on a fresh analysis of NSS household data wherein wages have been estimated for manual casual labourers. However, NSS household data on wages suffer from clear infirmities with a significant number of extreme values. We have therefore resorted to trimming the data set of 1 per cent of the extreme values at both the upper and lower ends.

    The resultant exercise shows that the growth rate of agricultural wages decelerated during 1993-94 to 1999-2000 compared to the period 1983 to 1993-94. Although there are state level variations, nine of the 15 major states experienced such a deceleration and two states showed negative growth rates during 1993-94 to 19992000. Noticeably, during the first period, states with low initial wages experienced much higher growth rates in agricultural wages, but this was not the case in the second period.

    Our results show that the growth rate of manual non-agricultural wages does not show a deceleration during the second period but seven of 15 states showed lower growth rate of real nonagricultural wages in the post-reform period. Like agricultural wages, non-agricultural wages for low wage rate states experienced a faster growth compared to the high wage rate states during the pre-reform period. The post-reform period saw a weakening of this trend. However, for non-agricultural wages, female wages show a higher growth rate in the post-reform period, at the national level and in seven of the 15 major states.

    For all casual workers, the all-India growth rates for male casual labour in both agriculture and non-agriculture (Tables 4 and 5) shows a slowdown in the second period. Total (male+female) growth rate in wages also showed a decline in the second period at the all-India level, although the decline in the case of nonagricultural wages was very modest.

    The interstate pattern of growth in wages was such that the coefficient of variation rose for both manual agricultural and nonagricultural wages between 1993-94 and 1999-2000. The gap between manual agricultural and non-agricultural wages also widened. However, the relative gender gap in wages improved slightly.

    Keeping these trends in mind, we have analysed the determinants of wages in the three time periods (1983, 1993-94 and 19992000) using panel data for 15 states. The four variables, which explain interstate variation in agricultural wages over the entire

    Appendix Tables

    Table A1: State-wise Level of Real (at 1999-2000 Prices) Rural Wages in Manual Agricultural Operations

    By Sex: 1983 to 1999-2000

    States Male Female Total 1983 1993-1999-1983 1993-1999-1983 1993-199994 2000 94 2000 94 2000

    AP 23.8 31.4 39.7 15.9 22.1 26.7 19.9 27.0 33.3 Assam 31.9 40.7 41.8 28.6 35.5 35.8 31.3 39.5 40.5 Bihar 20.0 27.4 35.4 16.4 23.6 32.1 19.0 26.3 34.5 Gujarat 28.2 31.8 39.1 25.2 30.7 34.2 27.0 31.4 37.2 Haryana 47.3 52.3 57.2 47.5 43.5 50.8 47.4 49.4 55.8 Karnataka 21.3 32.5 40.1 15.0 23.1 27.2 18.5 28.4 34.6 Kerala 54.6 63.2 91.2 38.1 46.7 61.9 49.4 58.7 82.9 MP 18.3 27.9 28.8 15.5 22.9 24.7 17.1 26.0 27.1 Maha

    rashtra 23.3 32.0 38.1 13.9 20.4 25.0 18.6 26.0 31.5 Orissa 17.9 27.6 29.4 13.4 20.1 23.0 16.5 25.2 27.2 Punjab 49.6 64.1 63.1 38.7 57.6 54.0 48.3 63.6 62.6 Rajasthan 33.4 43.4 50.5 22.4 35.3 38.1 29.2 40.8 46.9 TN 23.2 41.0 52.5 13.4 23.9 30.2 18.7 32.9 41.9 UP 25.0 34.3 39.6 19.9 25.9 29.9 23.8 32.5 36.8 WB 33.0 50.0 44.0 28.5 42.5 37.3 32.2 48.8 42.9 India 24.4 34.3 40.1 17.1 24.1 28.6 21.7 30.6 35.8

    Table A2: State-wise Level of Real (at 1999-2000 Prices) Rural Wages in Manual Non-Agricultural Operations

    By Sex: 1983 to 1999-2000

    States Male Female Total 1983 1993-1999-1983 1993-1999-1983 1993-199994 2000 94 2000 94 2000

    AP 27.8 36.8 50.6 19.8 24.0 29.8 24.9 32.3 46.9 Assam 37.5 50.7 50.3 23.3 31.9 33.1 35.3 47.2 48.7 Bihar 25.4 39.0 44.4 17.2 30.2 36.9 23.4 37.4 44.0 Gujarat 39.9 47.1 51.6 27.0 31.1 36.9 36.7 44.7 48.0 Haryana 45.7 58.1 67.7 33.0 33.8 56.5 45.4 57.0 67.0 Karnataka 30.6 43.8 59.2 15.6 27.4 36.7 25.1 38.9 54.8 Kerala 48.6 69.1 88.6 25.0 36.1 51.6 39.9 59.7 79.4 MP 27.9 36.0 40.1 20.2 25.2 31.9 25.8 33.0 38.1 Maha

    rashtra 27.7 43.7 54.3 16.9 25.8 39.3 23.5 39.3 51.8 Orissa 20.9 32.4 39.3 17.2 25.4 27.4 19.8 30.8 37.0 Punjab 50.7 63.5 68.1 28.5 31.2 51.2 48.8 59.0 66.5 Rajasthan 33.9 44.4 55.7 22.8 32.1 40.0 30.8 42.4 53.5 TN 27.4 45.1 69.2 13.4 25.1 34.4 22.8 40.0 62.1 UP 33.9 43.3 48.1 20.6 33.2 42.6 32.7 42.5 47.8 WB 35.1 55.3 48.1 20.3 37.7 32.2 32.3 51.0 46.1 India 30.6 44.9 55.4 18.3 28.0 37.7 27.3 41.3 52.7

    Table A3: State-wise Level of Real (at 1999-2000 Prices)for Casual Manual Work

    By Sex: 1983 to 1999-2000

    States Male Female Total 1983 1993-1999-1983 1993-1999-1983 1993-199994 2000 94 2000 94 2000

    AP 24.4 32.1 41.3 16.3 22.2 26.8 20.6 27.6 34.6 Assam 33.1 41.9 45.0 27.6 35.2 35.3 32.1 40.5 43.3 Bihar 20.6 28.4 36.6 16.5 23.9 32.2 19.4 27.2 35.5 Gujarat 29.9 35.0 42.0 25.3 30.7 34.6 28.2 33.5 39.4 Haryana 46.7 54.8 60.5 47.2 42.8 51.4 46.8 52.2 58.9 Karnataka 23.1 33.8 42.4 15.1 23.4 27.6 19.7 29.4 36.4 Kerala 52.9 65.2 89.8 33.6 43.0 56.8 46.4 59.0 81.1 MP 19.8 28.9 30.0 15.9 23.1 25.1 18.2 26.8 28.1 Maha

    rashtra 24.4 33.6 40.8 14.4 20.6 25.5 19.6 27.2 33.6 Orissa 18.3 28.3 30.7 13.9 20.6 23.3 17.0 26.0 28.2 Punjab 49.8 64.0 64.4 37.3 51.5 53.0 48.4 62.9 63.7 Rajasthan 33.6 44.0 53.8 22.6 34.0 38.9 29.9 41.7 50.7 TN 24.2 42.0 56.5 13.4 24.1 30.5 19.5 34.2 45.5 UP 27.4 36.5 42.0 19.9 26.6 30.6 25.9 34.7 39.3 WB 33.4 51.1 44.7 26.7 41.0 36.6 32.2 49.3 43.4 India 25.7 36.3 43.3 17.2 24.5 29.3 22.7 32.3 38.5

    Economic and Political Weekly September 23, 2006

    Table A4: State-wise Growth Rate of Real (at 1999-2000 Prices) Rural Wages for Casual Manual Work

    By Sex: 1983/1999-2000

    States Male Female Total 1983 1993-1999-1983 1993-1999-1983 1993-199994 2000 94 2000 94 2000

    AP 2.6 4.3 3.2 3.0 3.2 3.1 2.8 3.8 3.2 Assam 2.3 1.2 1.9 2.3 0.0 1.5 2.2 1.1 1.8 Bihar 3.1 4.3 3.5 3.6 5.0 4.1 3.3 4.6 3.7 Gujarat 1.5 3.1 2.1 1.8 2.0 1.9 1.7 2.7 2.1 Haryana 1.9 7.3 3.8 4.1 3.3 3.8 2.1 7.2 3.9 Karnataka 3.7 3.9 3.8 4.2 2.8 3.7 3.9 3.6 3.8 Kerala 2.0 5.5 3.3 2.4 4.8 3.2 2.3 5.4 3.4 MP 3.7 0.6 2.6 3.6 1.3 2.8 3.8 0.8 2.7 Maha

    rashtra 3.1 3.3 3.2 3.5 3.7 3.5 3.2 3.6 3.3 Orissa 4.2 1.3 3.2 3.9 2.0 3.2 4.2 1.4 3.1 Punjab 2.4 0.1 1.6 3.1 0.5 2.2 2.5 0.2 1.7 Rajasthan 2.6 3.4 2.9 4.0 2.3 3.4 3.2 3.3 3.3 TN 5.4 5.1 5.3 5.7 4.0 5.1 5.5 4.9 5.3 UP 2.8 2.4 2.6 2.8 2.4 2.6 2.8 2.1 2.6 WB 4.1 -2.2 1.8 4.2 -1.9 1.9 4.1 -2.1 1.8 India 3.3 3.0 3.2 3.4 3.0 3.3 3.4 3.0 3.2

    Table A5: State-wise Level of Real (at 1999-2000 Prices) Male Agricultural Wages

    By Sex: 1983/1999-2000

    States 1983 1993-94 1999-2000 NSS RLE AWI* NSS RLE AWI NSS RLE AWI

    Andhra

    Pradesh 23.8 16.4 30.9 31.4 25.4 43.7 39.7 30.9 45.1 Assam 31.9 26.9 36.5 40.7 35.5 47.4 41.8 41.3 48.9 Bihar 20.0 15.0 26.6 27.4 24.4 40.0 35.4 16.8 44.3 Gujarat 28.2 19.2 34.0 31.8 25.8 39.6 39.1 31.7 67.5 Haryana 47.3 23.3 63.7 52.3 30.1 83.6 57.2 46.4 102.9 Karnataka 21.3 13.5 25.8 32.5 25.1 34.9 40.1 31.9 42.8 Kerala 54.6 36.1 63.8 63.2 53.0 84.0 91.2 67.2 129.1 Madhya

    Pradesh 18.3 12.6 24.5 27.9 22.2 41.9 28.8 27.4 45.3 Maha

    rashtra 23.3 14.8 24.1 32.0 25.7 45.0 38.1 32.6 43.8 Orissa 17.9 11.1 19.5 27.6 21.0 38.3 29.4 27.1 35.3 Punjab 49.6 35.7 59.6 64.1 46.7 85.7 63.1 64.2 77.5 Rajasthan 33.4 20.0 51.6 43.4 27.8 49.4 50.5 42.7 69.4 Tamil

    Nadu 23.2 15.9 25.7 41.0 26.4 45.1 52.5 40.4 61.7 Uttar

    Pradesh 25.0 14.9 30.4 34.3 26.1 46.4 39.6 33.4 52.1 West

    Bengal 33.0 21.4 35.6 50.0 42.7 78.3 44.0 51.2 69.1 India 24.7 16.5 35.2 34.3 26.5 52.6 40.1 34.1 62.0

    Note: * The wage rate in case of AWI is for 1982-83.

    period, are worker-productivity, investment in agriculture, rural occupational diversification, and the percentage of agricultural labourers in the rural workforce.

    EPW

    Email: ravi@mail.jnu.ac.in

    Notes

    [The authors are grateful to P M Kulkarni and S R Raveendran for commentsand feedback.]

    1 Himanshu (2005) points out that the wage rates from NSS 43rd rounddoes not replicate and are not even comparable with the wage estimatesfor that round published by NSSO itself.

    2 Appendix Tables A1, A2 and A3 give the level of real wages in manualagricultural, non-agricultural operations and total rural wages for manualwork, respectively for the three points of time. Kerala, Punjab, Haryanahave high wage levels while Bihar, Maharashtra, Andhra Pradesh, Orissaand Madhya Pradesh have low wage levels. The wage rate for the highwage rate state is more than double than that of the low wage rate statein 1983 with this gap widening in the subsequent years. Similar trendsemerge for manual non-agricultural wages.

    3 Appendix Table A4 gives the growth rates for total rural wages for manualcasual labour. A deceleration in growth rate of rural wages is evident inthe post-reform period with growth rates declining from 3.3 per cent inthe pre-reform period to 3 per cent in the post-reform period for both malesand females. Eight states record a decline in the post-reform period withthe decline being sharpest in West Bengal, Punjab, Madhya Pradesh andOrissa.

    4 The growth rates are point-wise.

    5 Appendix Table A5 gives the level of male agricultural wages fromdifferent sources. Wage rates based on AWI are higher in comparison toNSS and RLE wage rates while NSS wage rates are higher than RLE wage rates.

    6 Development expenditure includes development expenditure under threeheads (i) agriculture and allied activities, (ii) irrigation and flood controland (iii) rural development. For other major heads of developmentalexpenditure on economic services such as special area programme, energy,industry, transport, communications, science and technology and othergeneral economic services it was not possible to differentiate the rural component.

    References

    Bardhan, P K (1984): ‘Determinants of Supply and Demand for Labour ina Poor Agrarian Economy: An Analysis of Household Survey Data fromRural West Bengal’ in H P Binswanger and M P Rosenzweig (eds),Contractual Arrangements Employment, and Wages in Rural LabourMarkets in Asia, pp 242-62.

    Bhalla, Sheila (1997): ‘Trends in Poverty, Wages and Employment in India’,The Indian Journal of Labour Economics, Vol 40, No 2, April-June,pp 213-23.

    – (1993): ‘Dynamics of Wage Determination and Employment Generationin Indian Agriculture’, The Indian Journal of Agricultural Economics, Vol 48, No 3, July-September, pp 448-70.

    Deaton, Angus and Jean Dreze (2002): ‘Poverty and Inequality in India: ARe-examination’, Economic and Political Weekly, September 7.

    Haque, T (1998): ‘Regional Trends Patterns and Determination of AgriculturalWages in India’, The Indian Journal of Labour Economics, Vol 41, No 4, pp 845-60.

    Himanshu (2005): ‘Wages in Rural India: Sources Trends and Comparability’,The Indian Journal of Labour Economics, Vol 48, No 2, April-June,pp 375-406.

    Lal, Deepak (1976): ‘Agricultural Growth, Real Wages, and the Rural Poorin India’, Economic and Political Weekly, Review of Agriculture, June,pp A47-A61.

    Panchamukhi, V R (2000): ‘Social Impact of Economic Reforms in India:A Critical Appraisal’, Economic and Political Weekly, Vol No 35, Issue No 10, p 836.

    Parthasarthy, G (1996): ‘Recent Trends in Wages and Employment ofAgricultural Labour’, The Indian Journal of Agricultural Economics, Vol 51, Nos 1 and 2, January-June, pp 145-167.

    Radhakrishna, R and C Ravi (2003): ‘Malnutrition in India: Trends andDeterminants’, Paper No 22, Centre for Economic and Social Studies,Hyderabad.

    Rao, V M (1972): ‘Agricultural Wages in India: A Reliability Analysis’, The Indian Journal of Agricultural Economics, Vol 48, No 3, July-September,pp 38-62.

    Rao, C H Hanumantha (1994): ‘Economic Reforms and the Prospectsfor Rural Labour’, The Indian Journal of Labour Economics, Vol 37, No 1.

    Sarmah, Sasank (2001): ‘Agricultural Wages in India: A Study of States andRegions’, The Indian Journal of Labour Economics, Vol 45, No 1, pp 89-116.

    Sen, Abhijit (1994): ‘Rural Labour Markets and Poverty’, The Indian Journal of Labour Economics, Vol 37, No 4, October-December, pp 575-608.

    Sharma, H R (2001): ‘Employment and Wage Earnings of AgriculturalLabourers: A State-wise Analysis’, The Indian Journal of Labour Economics, Vol 44, No 1, pp 27-38.

    Srivastava, Ravi and Richa Singh (2005): ‘Economic Reforms and AgriculturalWages in India’, The Indian Journal of Labour Economics, Vol 48, No 2, April-June, pp 407-24.

    Sundaram, K (2001a): ‘Employment and Poverty in 1990s Further Resultsfrom NSS 55th Round Employment: Unemployment Survey, 1999-2000’,Economic and Political Weekly, August 11, pp 3039-3326.

    – (2001b): ‘Employment-Unemployment Situation in the 1990s: Some Resultsfrom NSS 55th Round Survey’, Economic and Political Weekly,March 17-23, pp 931-40.

    Unni, Jeemol (1997): ‘Employment and Wages among Rural Labourers:Some Recent Trends’, Indian Journal of Agricultural Economics, Vol 52, No 1, January-March, pp 59-72.

    Dear reader,

    To continue reading, become a subscriber.

    Explore our attractive subscription offers.

    Click here

    Comments

    (-) Hide

    EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

    Back to Top