Why our GDP data can’t be trusted
Official growth and inflation figures are a mixture of guesswork and flawed accounting
Quarterly performance figures released on Friday showed the economy grew 0.8 per cent in the first three months, its slowest quarterly growth in four years ... The figure was down from the 1.93 per cent growth recorded in the previous quarter.
SCMP, May 16
Let me describe for you the exactitude of measurements of the growth of gross domestic product. They are like driving one of those old manual shift cars with three gears. Your have three forward speeds – zero to four per cent, four to eight per cent, and probable nonsense.
And don’t even try to put a number on it when you’re in reverse.
They are called national accounts, which may give you the impression that they are similar to audited corporate accounts. In fact, they comprise nowhere near as full a statement, or as precise.
Where a public company gives you a profit and loss account, a balance sheet, a cash flow statement and detailed notes to the accounts, these national accounts give you only a summary of the application of funds, which is the equivalent of one half of a corporate cash flow statement.
Among their many failings, they make no provisions for amortisation of fixed assets. For example, if a company buys a delivery van which it expects to last for five years, it will deduct a fifth of the cost of that van every year from its gross earnings and from the value of the fixed assets in its balance sheet.
A government does no such thing in its national accounts. It just treats its full expenditure on the van as an immediate increase in fixed capital formation and this goes straight into the year’s GDP growth number.
This matters because governments treat GDP growth as a form of report card and are prone to making wasteful expenditures to keep their grades up.
Resource rich countries, however, would not encourage quite such devil-may-care development of their mineral resources if they had to deduct the value of the minerals extracted in their territories from GDP every year, as is done by the mining companies in their own corporate accounts.
They would also think twice about spending money for economic stimulus if proper accounting showed them they were getting nowhere near as much real stimulus as they had expected.
Admittedly our own Hong Kong government now publishes accruals accounts that make provision for these matters but it is the cash accounts on which this 0.8 per cent quarterly growth was based. The accruals accounts are only published later and then no-one except my friend Stephen Brown pays them any regard.
Worst of all, GDP numbers come in two forms – dollar of the day and inflation adjusted. That 0.8 per cent figure in the first quarter was the inflation adjusted one. The dollar of the day growth figure was 3.1 per cent, implying an inflation rate of 2.3 per cent.
This is all very well except that inflation here is calculated for the entire economy, not just the consumer price basket, and incorporates almost as much guesswork as the Mark 6 lottery.
In our case, in particular, these overall inflation figures can be thrown wildly one way or the other in the net balance of foreign trade.
Question: Just how can anyone truly come to an accurate figure of average inflation across a million items of exports and imports? Answer: You lick your finger and stick it into the air.
There is also the occasional suspicion that the dollar of the day figures are doctored to make the inflation adjusted ones look better. The mainland’s 6.9 per cent inflation adjusted growth last year, for instance, was achieved only on the assumption that the entire economy was in deflation during the year. Perhaps.
Let’s cut to the chase. There are good times and there are bad times and, on balance, it looks like we are probably headed for a batch of bad times at present.
This puts all economic performance measures into question. When our chief budgeteer, Whiskers John, then tries to cheer you up with talk of “positive signals”, just remember that when this car is in reverse you can never really tell just how fast or where it is going.