Coen Welsh | Sep 20, 2017 | 0
Nominal GDP is perhaps the only reliable growth measure
Trying to determine accurately a country’s gross domestic product is an arduous undertaking, even for developed economies where mountains of raw data and some fancy statistics are readily available.
To determine GDP for a least developed country is almost impossible. The structures, institutions and capacity needed to collect data, develop models and massage the statistics to align them with reality, simply do not exist.
Measuring the GDP for a middle income country is not difficult but neither is it easy or straightforward. Take us for instance. The data requirements are immense. Information has to be gathered from many agencies over the entire reporting period across all the sectors from all the regions. And this not only applies to government ministries, it also applies to all private sector companies who operate nationally.
Income or revenue derives overwhelmingly from taxes. The three major components are taxes on income and profits, taxes on consumption, and taxes on international trade. We know these as PAYE and company tax, VAT, and SACU transfers. Although somewhat delayed, these statistics offer a very clear picture of the revenue side in any given fiscal year. These figures are also fairly reliable since there are several mechanisms in place that test their validity. There is also a strong correlation between revenue and the state of the domestic economy.
In essence, it means that as far as budgeting goes, both the expenditure side and the revenue side should provide us with statistics that are more than 95% reliable. With a standard deviation of less than 5%, this would generally be accepted by any statistician as ‘reliable’.
The difficulty arises when one wants to calculate certain ratios, particularly those that express the sub-components relative to the size of the revenue stream of the whole economy, commonly called gross domestic product.
It was not my intention to open a can of worms regarding the protocol to measure GDP, but it has certainly stirred the hornets’ nest.
Accurately measuring, or capturing GDP for a middle income country, carries indeed many risks, most of them on the downside. GDP is suppose to capture ALL receipts by All economic players. This includes the state, the private sector, and the informal sector. To capture government revenue is easy, the Receiver does that for us. To capture all private sector income is more difficult since there are many players who do not pay tax, and there are many concessions, especially in the case where a company or a commercial entity has made a loss. To capture the contribution from the informal sector is impossible. This can only be extrapolated, based on a very few points of discovery, for instance where informal sector players keep bank accounts. But this is only the minority.
What this boils down to is that our GDP is an educated guess, at most. To make matters worse, reliable GDP figures are only available long after the end of the reporting period. As more and more data flow gets captured into the reporting model, GDP is revised, sometimes even three years after the actual period.
Futhermore, we are most dependent on accurate GDP figures for budget purposes but this entire body of documentation is forward looking, in other words, it pertains to the future. Again, as any financial planner knows, this is extremely difficult to predict regardless of the framework.
In my mind, the only reliable indicator is nominal GDP since this is made up of all receipts by all entities. Even in our situation where reporting is often scanty or non-existent (informal sector), the available sum total of receipts for known sectors can be collated and extrapolated for the economy as a whole. These are then corroborated with other statistics, and we get a picture that we regard as reliable or sufficiently reliable for us to use in our economic planning.
But nominal GDP naturally falls into two sub-components. First is the change in the volume of goods (and services) and secondly is the change in the value of goods (or services). This last one is indicated by inflation, or at least the impact inflation has on purchasing power. This implies that in any GDP, there is an automatic deflater. However, to determine the exact value of the deflater is even more risky than to claim to have accurate GDP figures. And then one has to realise there is more than one methodology to apply the deflater to nominal GDP. Taken together, this inherent uncertainties built into the system should make us very circumspect when we deal with macro-economic statistics.