Labour force survey on a quarterly basis was introduced in Bangladesh for the first time in 2015, and July-September 2015 was the first quarter to be covered by the survey. The results of that survey, released earlier this year, naturally captured the attention of users of all kinds, ranging from media to researchers and policy makers. One point that was highlighted by many is the much slower growth of employment between 2013 and 2015 compared to the earlier inter-survey period of 2010 to 2013. And based on that, one worried that the employment generating capacity of the economy has declined considerably. But is that really so?
If one looks at numbers alone without considering the methodologies and procedures used in the surveys of various years, one may indeed be led to such a conclusion. But that would not be appropriate. Before getting into such details, let's take another look at some numbers.
Total employed population (over 15 years of age) in the country was reported as 58.1 million in 2013 and 58.7 million during the July-September period of 2015. From these two figures, it might seem that in two years from 2013, employment increased by a figure of 600,000 (300,000 per year). On the other hand, between 2010 and 2013, the number had risen from 54.1 million to 58.1 million - an increase of 4.0 million in three years - 1.3 million per year, which is much higher than 300,000, the figure for 2013 to July-September of 2015. From such comparison - if one could make it - it would appear that employment growth has declined drastically. But the key question is: would such a comparison be valid? The answer is: No! In order to understand why the comparison would not be valid, it would be necessary to look at the methods and procedures used in the labour force surveys conducted in various years.
Consider, first, the Quarterly Labour Force Survey (QLFS) of July-September 2015. Data from this survey would be for that period rather than for 2015 as a whole. Can this make a difference for the employment figure or other such aspects of the labour force? The answer is likely to be negative for regular employment where there is no seasonal variation. However, a large part of the economy of Bangladesh, especially activities relating to agriculture, is still characterised by seasonal variation - although the degree and nature of such variation has changed over time. Hence the same person may come up with different answers about her employment status depending on the time of the year when she is asked the question.
As for seasonal peaks and troughs of agriculture and allied rural activities, July, August and September are not among the busy periods. The boro crop would have been harvested before this period. The other major crop, viz., aman rice is usually planted in September; and hence there may be a short busy period during that month. But on the whole, the three months mentioned above would perhaps be one of the leaner periods in the crop cycle, and hence, the employment figure for that period (as reported in July-September QLFS) maybe more of a reflection of that rather than a real decline.
On the other hand, the survey of 2013 was carried out throughout the year, with a fixed number of sample households visited every month. As a result, the impact of seasonal variation in economic activities is most likely to have been taken care of. The figures based on 2013 survey can be said to reflect the situation in 2013 as a whole, not for any particular month or quarter.
Given the difference in procedures for carrying out the survey of 2013 and the QLFS of 2015, the estimates based on them are not strictly comparable. In the circumstances, how can one address the question as to what has happened to growth of employment (or for that matter, other variables like economically active population, etc.)? One way would be to wait till three other QLFS - viz., for October-December 2015, January-March 2016, and April-June 2016 - have been completed and the results from all four quarters are combined to produce estimates for July 2015 to June 2016. At that point, the 2015-16 figures could be compared to those for 2013. Another way of having comparable figures would be to carry out a QLFS for July-September 2016 and compare the results from that with those of July-September 2015.
In the absence of comparable data from labour force surveys, is there any other way of addressing the question of what has happened to employment growth in recent years (i.e., after 2013)? Outside agriculture, a major source of employment is manufacturing, and within that sector, the ready-made garment (RMG) industry is a major source of employment. The Bangladesh Garment Manufacturers and Employers Association (BGMEA) provides data on key aspects of the industry including exports and employment. Data relating to employment, exports and number of factories (presented in the Table below) indicate a few interesting aspects relating to growth and employment in the industry.
If BGMEA data on employment are to be taken seriously, it would appear that total employment in the sector has remained unchanged (at 4.0 million) since 2011-12. If true, this would raise the question whether the sector has reached some kind of limit regarding employment expansion. Another interesting piece of statistic is the number of factories: the total number peaked at 5,876 in 2012-13, and since then has declined to 4,328 in 2015-16. These two indicators, viz., the number of factories and employment might create an impression that the industry has stopped growing. Fortunately, however, that is not the case. As is indicated by export figures, the industry is still growing at healthy rates. What, then, is happening in the industry, especially regarding employment? While serious investigation is needed on various aspects of the industry in order to address this question, a few remarks by way of conjecture may be made.
First, it is quite possible that the employment figure has simply not been updated, and hence does not provide a correct picture of what has been happening in recent years. But if the employment figure is taken seriously, that, combined with the export figure would imply that growth of output in the sector has taken place without any expansion in employment. Another piece of information provides a pointer in the same direction: the decline in the number of factories. It is possible that the industry is undergoing some kind of structural change that is resulting in the closure of some factories, especially smaller ones - which may also happen to be the more labour-intensive ones. Such a conclusion may have some basis if one considers the developments in the sector since the Rana Plaza disaster of 2013, especially a stronger focus on workplace safety issues and the associated pressure for upgrading factories to meet the basic safety standards. It is possible that some factories have closed down in the wake of that disaster, leading to loss of jobs. And yet, exports from the sector has kept rising, thanks to the ability of the larger enterprises to meet growing demands. If that has been the case, there would be a serious question on the ability of the sector to continuously expand employment and absorb the surplus labour available in the country.
If employment growth in the RMG industry has declined in recent years (the likelihood of which has been indicated above), it is quite possible that overall employment growth in the country has also declined. However, given the problems involved in comparing data from the labour force surveys of 2013 and July-September 2015, and the question about the reliability of the recent employment figures provided by BGMEA, all one can say at this point is that the jury is just out, and one has to wait for the verdict.
The author, an economist, is former Special Adviser, Employment Sector, International Labour Office, Geneva. [email protected]