Off the markIt ain’t so

One of the most difficult problems that economists encounter in analysing economic problems of Bangladesh is the paucity of appropriate data. This is exacerbated by the fact that the available data are often of poor quality. Any analysis done with inadequate and inaccurate data is in danger of coming out with incorrect findings and conclusions. As the sole supplier of aggregative data, the government should take special care in providing adequate and accurate data in real time to ensure correct analyses and appropriate policy responses.

The use of incorrect data for economic analysis, and in particular for policy-making, can be costly. Once incorrect data are released to the public domain, many people including government functionaries will use them to arrive at conclusions and decisions that could be erroneous. This may be illustrated by a recent event. The agriculture ministry claimed that there was a bumper aman crop harvest of 13.1 million tonnes in 2009 (against 11.6 million tonnes in 2008). Such a large harvest should have guaranteed adequate supply of rice to the market for at least the first three months of 2010, and prices should have been stable or falling.

Surprisingly rice prices started increasing soon after the harvest. Caught by the unexpected price increase, the kneejerk reaction of the government and the civil society was to blame the proverbial 'unscrupulous' traders for the price hike. They were accused of syndication and hoarding despite their repeated complaint that sufficient rice was not coming to the market. It was not explained how a large number of rice traders of modest means in different parts of the country could form and maintain a cartel, nor was any evidence of hoarding made public. It was also not clear why hoarding should be a concern with such a bumper crop. With an actual bumper output of potato the main problem was not being able to hoard!

Economists were at a loss to explain why the prices were rising, especially when the international prices of both rice and wheat were on a downward trend at the end of 2009. Some succumbed to the comfort of the 'conventional wisdom' of the syndicate theory, while some others hypothesised that farmers were hoarding their produce instead of releasing to the market in anticipation of higher prices in future. Each proffered explanation seemed to beg more questions than they answered.

There is an oft-quoted piece of old wisdom eponymously named the Occam's Law which states that 'Entia non sunt multiplicanda prater necessitatum', ie 'Entities [of explanation] should not be multiplied beyond necessity'. Put in simple terms, the law asserts that with any problem 'all things being equal, the simplest solution is the best one'. When applied to a natural or social problem, it does not imply that the real situation is not complex, but that the explanation that requires the fewest assumptions is most probably the correct explanation.

Applied to the rice price hike problem after the aman harvest, the simplest answer requiring the minimum number of assumptions was of course a shortage of rice (or excess demand) in the market. The ordinary law of the market then explained the price hike. The problem, however, was that the agriculture ministry precluded this answer by claiming a record harvest of the aman crop. Consequently, people sought answer to the conundrum with additional assumptions such as syndication and hoarding.

The food directorate, food and disaster management ministry indirectly contradicted the claim of bumper harvest in a report by suggesting that the sharp increase in the price of rice when the agriculture division estimated 13 million tonnes of aman harvest was simply 'unprecedented'. The very first reason it provided for the price hike was that the aman harvest was considerably lower than expected. In other words, the agriculture ministry had substantially overestimated the harvest. Bangladesh Bureau of Statistics estimated aman output at 12.2 million ton, while World Food Programme put the figure at 11.5 million tonnes. The report leaves little room for doubt about a real supply shortage in the rice market. It goes on to list a host of other reasons, such as a switch to low-yield fine varieties of rice, export bans by India, hoarding by households (not traders!), extreme cold and fog during the winter etc. that contributed to the supply shortage. It also cites a demand factor, the recent increase in the salaries of public employees, as a possible reason for the upward movement in the rice price. All of these explanations imply a relative supply shortage (or excess demand) in the rice market.

The price hike, that has caused misery of the ordinary people and embarrassed the government (particularly those government high-ups who blamed the business people for unethical practices), could have been avoided if the agriculture ministry did not put out overestimated data about the aman harvest. The government was ensnared in the web of the erroneous claim that eventually had unpleasant ramifications for the economy.

That the government was taken by the hype of a bumper harvest is seemingly suggested by the cessation of import of any rice by the government of its own accord; the only import of a measly 3631 tonnes in the public sector in 2009-10 (up to March 2010) was financed by foreign aid. In contrast, the government had imported 415,304 tonnes of rice during FY2008-09 despite the fact that the international price was the highest of the last forty years. There was also a near shutdown of the government procurement machinery during the aman season. The government set a low procurement price of Tk 22 for aman rice based perhaps on the then comfortable rice market situation and the expectation of an imminent bumper harvest. It is arguable that it perhaps wouldn't have set such a procurement price if it had anticipated a poor harvest. The market price of rice had already started rising by the middle of December 2009. The procurement price turned out to be too low, which dissuaded farmers from selling to the government. The total amount of aman procured by the government stood at a negligible 14,690 tonnes (as of March 31, 2010) against a target of 302,700 tonnes, ie, less than 0.5 percent of the target. The rice stock fell sharply from 1.2 million tonnes before aman harvest to 0.3 million tonnes by March 2010. During the same period the previous year the rice stock had decreased only marginally from 1.2 million tonnes to 1.1 million tonnes.

The bumper crop hype must have also misled the private sector rice importers. They virtually ceased rice importation. While they had imported 182,718 tonnes of rice during July-March 2008-09, they imported only 23,756 tonnes during July-March 2009-10, ie, only 1.3 percent of the previous year's import. Wheat import was not negatively affected; it actually increased from 1.3 million tonnes in July-March 2008-09 to 2.3 million tonnes in July-March 2009-10. This ensured adequate supply of wheat in the market, and hence the wheat price remained stable.

If the possibility of an ordinary aman harvest were flagged early in the season, both the public and the private sector might have been activated in anticipation of a relative supply shortage. This would have helped to keep a lead on the rice price. Overestimated data could have misled the market into inaction when there was a need for import. In the end, the rice price increased in the domestic market despite the fact that the price was on a downward trend in the international market. Government and civil society 'experts' predictably blamed the rice traders.