Tuesday 17 September 2013

The case for data indices

This article in published in the "money matters" on 16 Sep 2013...

The analysis of an event requires the use of logical facts that are based on observational features. That is the art of conducting an effective analysis. Meanwhile, it can also be based on data.  Data helps reveal accurate and logical results. Hence, it plays a crucial role in analysis. Different data sets are available for different countries in the world that have their own definitions and adjustment factors. For instance, the GDP data obtained from Pakistani sources will differ from that obtained from World Bank and IMF publications or the CIA Factbook. Although international sources obtain data from Pakistani sources, it is then changed based on their own definitions.

Besides data obtained from local sources, international organisations are also involved in collecting primary data. Transparency International and Global Competitiveness Index draw data from an opinion survey that is designed by them. On the other hand, some international sources obtain data from various local sources, including national accounts, Pakistan Bureau of Statistics and non-governmental organisations such as Gallup. Subsequently, an index is compiled using the country data, which also helps to compare that particular indicator with other countries.

Due to the nature of different data sets, definitions of variables and dissimilarities in the classification of fiscal year for each country, the indices could be misleading and thus the reliability of data sets comes under question. The problem lies with all those data sets (human development index, failed state index, etc.) that use macro variables, such as per capita income. Similarly, Transparency International’s corruption data is a perception survey but is generally considered as actual data on corruption. Further, the International Country Risk Guide’s quality of governance data represents points assigned to different variables by respondents in different countries.

Chances of the respondents being biased in attributing scores for various indicators in different countries are significantly high.

Recently, the Walt Disney Company decided to cancel textile import orders worth $150 million with Pakistan due to the country’s inability to improve its position in the worldwide governance index (WGI).

The cancellation of import orders based on specific data that has nothing to do with the quality of products is a non-tariff barrier on the textile sector by Walt Disney. It is anticipated that other firms will soon follow suit, which could cost the country more than $1 billion exports.

The WGI compiles aggregate and individual governance indicators for 215 countries using six dimensions of governance (voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rule of law and control of corruption) over the period 1996-2011. The index value for each of the six governance indicators varies between -2.5 and 2.5 (from worst to the best). Various data sources, including the Economic Intelligence Unit, Afrobarometer, etc., are used to develop these indices. More than 20 sources are used to construct each variable index and then the six variables are combined to construct the WGI.

While the data provides an insight into the overall performance for the last many years, it also portrays an astonishing economic picture. For example, the variables ‘political stability’ and ‘absence of violence’ have only two dimensions, i.e. political stability and absence of violence. In the last era of democratic government we had political stability but violence in terms of drone attacks and suicide attacks had increased. Thus respondents could give biased opinion if both the questions are asked together.

The data on political stability and absence of violence reveals that there was more political stability and lack of violence over the period of 1996-2000. Further, political instability and violence had risen from 2008 to 2011 as compared to the Musharraf/PML-Q era. Undoubtedly, drone attacks and acts of terrorism were more frequent during 2008-2011 but that era was the most politically stable. Nevertheless, the data tells us a different story, which raises questions about its authenticity.

Apart from voice and accountability, all the variables have shown deterioration in the last era of the PPP government. The WGI data might be misleading but it can be used effectively for the country.

All the six indicators show that the country’s position has remained below zero, which implies weak governance and the need to undertake strong measures to fix these issues. We need to stop seeking donors’ assistance and instead undertake soft reforms in different sectors, particularly civil service. Undertaking these reforms does not require heavy investment funds, rather strong will of the government and bureaucracy.

The cancellation of import orders by Walt Disney is a wake up call for the government. There is a possibility that Pakistan’s products could lose a substantial chunk of exports in future if companies look toward the WGI or any other indicator before entering into a trade agreement with the country. As the data represents the scenario till 2011, the question is how a well-established company could cancel a $150 million deal in 2013.
- See more at: http://magazine.thenews.com.pk/mag/moneymatter_detail.asp?id=6158&magId=10&catId=177#sthash.wZFsa8yL.dpuf

No comments: