Exports incentives linkage - Did DEPB matter?

This blog has emphasized many times about the importance of establishing the linkages between exports performance to incentives provided by the Government. The urgency of this hit me when I went to a customs commissionerate on an attachment. The perspective on export incentives, when seen from the revenue angle, is different to, when seen from export promotion angle. The revenue collectors look at such schemes as a loss in revenue. During one such talk and presentation given by a senior customs official, the officer said that removal of DEPB scheme had no impact on exports. Upon prodding, he came up with numbers and showed that the exports had remained same, or increased in some areas, after the DEPB scheme was withdrawn. So, he concluded, DEPB had no impact on exports performance. 

DEPB was an export incentive scheme launched with an intention of offsetting the duties incurred on the inputs that go into export products. Drawback is another such scheme, with similar intent of duty nullification on inputs. However, DEPB incentives were more lucrative, and in some cases, they were difficult to justify from revenue point of view. WTO objections and other issues led to withdrawal of this export incentive scheme last year. Drawback has stayed because the rates of incentives offered are modest and can be justified as genuine duty nullification. 

Coming back, I wondered about the statement. Cusps are interesting points for data analysis. When a scheme is introduced, or when a scheme is withdrawn, the zone of such changes, I call it a policy cusp. At these points or zones (with probably a time lag effect which can be taken into account), we can see interesting things happening. Policy impacts can be studied at these zones. However, cusps can also be misleading if read wrong. Like in the case above. The officer had not taken a lot of things into account.

If a proper policy cusp study is to be done for impact of an incentive scheme, one has to take into account a few things. First, one has to see the trend before and after the cusp. For example, in this case, the exports were probably growing at a rate of X%. After DEPB was withdrawn, the exports didn't decrease, but the rate might have changed. Exports, in terms of value, not decreasing doesn't say anything about change in rate. In fact, contrary to what the officer felt, the rate has indeed changed after DEPB withdrawl. The exports growth rate has drastically come down. Unfortunately, there is a confounding of that variable (DEPB) with global demand slump and due to this mixing, we cannot isolate the effect of DEPB easily. Also, the data change might have a time lag before it shows the effects. It might be more than a year, since DEPB licences are still floating and the benefits are accruing. In addition to that, export orders are mostly taken in advance which also adds to the lag effect. There might also be sectoral effects which do not show up when we take aggregate figures.

Such and other factors make drawing general judgement very risky. At the same time, from the export promotion point of view, it is foolish to say that increasing x% of incentives for exports, is bound to improve export performance. There are many factors, including the ones outlined above, to be taken into account before that statement is made.

Cusps are the zones that must be watched out for. However, we need very good data collection and analysis system to do good research. There is a strong push required to establish a rigorous methodology of data analysis and linking applied policy making with such analysis. Lobbies can always make their demands as long as the in house research of the department supports the claim.

To conclude, did DEPB matter? The honest answer is, I don't know. But that doesn't mean, it didn't matter, either. 


  1. I appreciate your idea of Cusps letting us in into the export data analysis as far as the schemes are concerned. But unless a commodity wise analysis of the data is done in a time series keeping into account the other variables and controlling them in the statistical analysis confounding the numbers, nothing much can be said about anything. Our training is really short on analysing the trade data statistics!!


Post a Comment

Comments are moderated. Your comment will be online shortly. Kindly excuse the lag