The rising unemployment rate seems to have been reversed a couple of years ago, going by data available from the National Sample Survey.
While this is based on 'thin' samples from the NSS, and therefore not completely reliable, several experts believe that the sample gets the trend correct. NSS data show that employment growth doubled, from 1.07 per cent between 1993-94 and 1999-00, to 2 per cent between 1999-00 and Oct 2002.
Analysis of employment trends is complicated because no two experts seem to agree on the quantum of the decline in job growth from the pre-reforms period (1983 to 1993-94) to the post-reforms period.
The NSS provides data on three types of employment -- usual status, weekly status and daily status. While the 'usual status' data shows there was 2.81 per cent unemployment in the country in 1999-00, 'weekly status' data puts it at 4.41 per cent and 'daily status' says it is 7.32 per cent.
The Montek Singh Ahluwalia Task Force in 2001 chose to use the 'usual status' data, while the S P Gupta Special Group felt 'daily status' was a better measure.
Economist Surjit Bhalla prefers to use 'weekly status' in keeping with the international norm, but gets different growth rates once he adjusts the NSS figures to take into account population figures, as enumerated by the census of 2001!
While all experts agree that employment growth fell dramatically during the second half of the 1990s, Bhalla says the decline looks less serious when you factor in the equally sharp decline in the number of people coming into the labour force, a result of the population spending more years getting educated instead of looking for jobs.
According to the Ahluwalia task force, in the period 1983-1994, the labour force grew by 2.05 per cent per annum while employment was just marginally lower at 2.04 -- in other words, employment growth was sufficient to absorb new job seekers.
In the 1994-2000 period, while employment growth fell to just 0.98 per cent, labour force growth also fell to 1.03 per cent.
Gupta's Special Group in contrast shows employment growth plunging from 2.7 per cent to 1.07 per cent between the two periods, and shows the growth of the labour force falling from 2.43 per cent to 1.31 per cent.
But since, in this case, employment growth fell faster than the decline in labour force, unemployment rates rose from 5.99 per cent in 1993-94 to 7.32 per cent in 1999-00. Bhalla's analysis shows no such disaster, with the fall in both categories broadly similar.
If unemployment is rising fast, it is logical that wages should be crashing. Yet, data culled by Bhalla from the Planning Commission's poverty figures show that wage rates went up by 10.7 per cent annually in the 1983-94 period, and then rose further by 12.5 per cent annually in the 1994-2000 period.
Adjust for inflation and the 'real' annual hike in wage rates was 2.4 per cent in the first period, versus 4.4 per cent in the second, a trend that is not consistent with increasing unemployment.
While Gupta argues that the impact of increased wages is offset by job losses, poverty expert K. Sundaram argues that the real wage growth was strong enough to offset a reduction in the number of days worked, and that there has been an improvement in the employment situation.
Suresh Tendulkar, another employment expert argues that the decline in employment in 1998 and 1999, in the factory sector for instance, is largely due to the sharp fall in industrial growth; in the 1993-96 period, when industry was booming, factory employment rose by 3 per cent each year. In other words, once growth picks up, so will employment.
More from rediff