Dear Editor,

Please allow me to congratulate Mr A Donald Augustin for introducing a very pertinent and intellectually stimulating discussion (‘Recommendations have been made to introduce a method of measuring GDP’ SN, September 26).

I heard the name A Donald Augustin previously, while attending UG, as someone who excelled as an economist in Guyana. If I recall correctly he was at the State Planning Secretariat or Ministry of Finance…never did get to meet him.

This is a very timely piece and a very good discussion away from the run of the mill discussions. I ran through the report over the weekend and my initial thoughts were as follows:

● this attempt is to measure happiness – recognizing the practical limitations to measurement;

● the discussion on national accounts, in particular, using consumption and income and disaggregated GDP measures, is timely particularly given the importance of balance sheet concepts through economic cycles;

● that this new approach would include contributions (final goods and services) which were previously excluded because no market value was attached to labour although it had significant value to the economy, eg, domestic housework by an unemployed spouse or a paid domestic worker; and

● whether this approach will address the per capita conundrum where it stands to reason that each and everyone has contributed equally to the national income despite the fact my father is a goldsmith and yours is a subsistence farmer (paraphasing Walter Rodney) in the same geographical place.

Finally, in reference the per capita conundrum, it doesn’t address distributional issues in a national accounts sense per se, but Stiglitz and Sen make the argument that more attention ought to be paid to income and wealth distribution. I don’t think it is possible to adjust for distribution without imposing a loss function which necessarily implies some sort of qualitative measure that isn’t neutral, primarily because it embodies some norm about what constitutes a fair distribution. See for example genuine progress indicator. Pure measures like Gini or Theil are discussed as dimensions of distribution but the real issue here is how do we come up with the aggregate. From a practical point of view, I think the approach should be to use multiple measures/concepts and apply deductive reasoning given the limited information, so we can capture each sector’s contribution – but then again we might run into a standardization problem.

Let the discussion continue.

Yours faithfully,
Evan Thomas