VR Logo

Unreal GDP vs real data

Don't pay any attention to the GDP numbers that are being trotted out around the world

Unreal GDP vs real data

'Japan's GDP shrinks at 27.8 per cent.' 'Singapore Slumps Into Recession With Record 41.2% GDP Plunge.' These are only two choice headlines that portray a grim picture of the state of the world economy. A 27.8 percent or a 41.2 percent decline in GDP is truly disastrous. Or is there more to the story than meets the eye? Actually, there's less to it. The reality is that these numbers are basically fake news. They are clickbait, which cynically twist news items to create headlines that will induce you into a panic-driven urge to read the article itself, thereby earning the publisher some pennies.

So what are the facts here, why am I saying that these numbers are fake news? Let's take Singapore in detail. In the April-June 2020 quarter, Singapore's GDP was measured to be lower than that in the April-June 2019 quarter by 12.6 percent. Because of the annual rhythm of economic activity, GDP (and so many other business and economic numbers) are best compared to the equivalent period of the previous year.

So where did that claim of 41.2 percent decline come from? The April-June quarter this year was lower than the immediately preceding quarter (January-March) by 12.2 per cent. The 41.2 was obtained by assuming that in each of the successive three quarters, the GDP figure would be 12.2 percent lower than the previous quarter. The fall would continue at the exact same pace, month after month, quarter after quarter.

If you ask those who are peddling the panic, they would say that such data should be annualised to understand the true impact. However, in the current situation, this is simply untenable. You annualise data when there is some reasonable basis to assume that the pace of change is stable. Moreover, the headlines do not even mention that this number is an annualised projection - they just say that Singapore's economy declined by 41.2 percent. Annualisation is a future projection based on a crude assumption. Even in the best of times, it is nowhere near accurate. To present it as a fact, as something that has already happened, just amounts to a false statement.

In any case, GDP itself is not a physical measurement in any meaningful sense. There's no such thing as GDP out there in the real world. The number itself is a result of a model that has a huge number of assumptions, heuristics and thumb rules built into it. The best you can say about GDP is that if the basis of these assumptions etc do not change, and the same methods of measurements continue to be used, then it's comparable to its own past measurements. As should be obvious, many significant assumptions about economies have been hit for a six over the last few months. Of all the things that you can use to measure the world today, GDP is the least useful. In fact, it's less than that - it's actively harmful to take it seriously and to make it part of your decision-making framework. Apart from generating some pennies of revenue for news websites when you click on a GDP story, it has no use at this point of time.

So what are the kinds of numbers that one should look at to figure out the scale of the economic impact of the Chinese virus? The answer is obvious - hard numbers that are actual measurements of some activity. The revenue that businesses are generating. The taxes that the government is collecting. The actual numbers of cars, shoes, TVs, petrol, diesel and shirts that people are buying. It takes effort to make sense of these kinds of numbers, and figure out what they mean for your savings and investments, but unlike GDP headlines, they do have the advantage of being real.