Singapore’s Volatile Data Stump Professionals

September 4, 2013, 5:32 AM

Singapore’s Volatile Data Stump Professionals

By Gaurav Raghuvanshi

Last month, Singapore’s government announced the economy grew 3.8% on-year in the second quarter. But as late as June, economists polled by the city-state’s central bank were predicting growth of just 1.5%. Economists got it wrong on exports too: They predicted a nearly flat print in the second quarter, when exports actually fell 5.0%. The difference was even starker in the first quarter: Economists in March predicted exports would fall 0.5%, but in fact they shrank a whopping 12.5%. The Monetary Authority of Singapore polls economists at banks and research firms every quarter on key local data such as gross domestic product, exports, currency, inflation and employment. The results are released at the start of every quarter, with the second-quarter survey landing Wednesday. It turns out that the 20 or so economists who respond to the survey get it quite wrong, quite often.Economic predictions are never easy. But they become even more complex in tiny Singapore, where trade is more than three times the size of GDP.

“Singapore being such a trade-dependent economy, one is effectively making predictions for key trading partners,” said Robert Prior-Wandesforde, head of Indian and Southeast Asian economics at Credit Suisse in Singapore.

According to Wednesday’s survey, the 19 economists now expect GDP to grow 4.0% on-year in the third quarter, up from the 3.5% they forecast in June. They’ve also raised their median estimate for full-year GDP growth to 2.9% from 2.3%.

Exports, however, could be a drag: The economists now expect exports to contract 0.5%, a sharp change from the 2.5% expansion they predicted in June.

With the next survey not due until December, only time will tell how accurate the new forecasts are. But it’s worth bearing in mind that Singapore’s data can get swung around by factors that don’t lend themselves to easy prediction.

For example, output of pharmaceutical drugs – which has a 15% weighting in total manufacturing output — can jump as much as 400% in a month when the small number of Singapore-based factories of large multinational firms produce batches of high-value drugs.

That indicator also can swing deep into negative territory when plants close for up to six weeks for maintenance between different batches of drugs.

The government itself often has to make big revisions as new information comes in. That means the data keep changing and, sometimes, end up resembling the economist forecasts much more closely than initially thought.

A comparison of forecasts and actual outcomes over the past four quarters shows that while economists were quite good at predicting the inflation rate and exchange rate, their growth and export projections were well off the mark.

In June, the economists predicted a U.S. dollar would be equal to 1.2600 Singapore dollars by the end of the month – not far off the actual rate of S$1.2650. Similarly, Wednesday’s report showed the average inflation rate for the second quarter was 1.6%, compared with the 1.8% economists predicted in June.

In the prior survey, economists predicted the U.S. dollar would be near S$1.2400 by the end of March, close to the actual rate of S$1.2440. At 4.0%, average inflation for the first quarter was close to economists’ median prediction of 3.9%. And the economists were right on the money about the unemployment rate at the end of the first quarter (1.9%).

“Economic forecasting is an art. If you could pinpoint the exact number, there wouldn’t be any need of a forecast,” said Selena Ling, head of treasury research at OCBC Bank.

While inflation is reported every month and economists can see the exchange rates on a real-time basis, quarterly data such as GDP are more difficult to predict because of external uncertainties.

Events in big trading partners such as the U.S., Europe and China — and even neighbors such as Malaysia, which held elections earlier this year – can have an outsized effect on Singapore’s data.

“What may have caught people off guard was the speed at which the U.S. came out of recession and China slowed more than expected,” Ms. Ling said. “At the end of the day, we are a small but open economy.”

Economists don’t have too much information on the services sector in Singapore. While purchasing managers’ indexes and monthly output data can give them a sense of industrial activity, no comparable indicators for the services sector are available, Ms. Ling and Mr. Prior-Wandesforde said.

Mr. Prior-Wandesforde also let out an industry secret: Economists don’t lose much sleep over whether these forecasts turn out to be right.

“At the end of the day, they are historic numbers. We would rather spend more time generating path-breaking ideas for the future. We are more interested in trends rather than the monthly or quarterly noise” that the data generate, he said. “In fact, our job is more to cut through that noise.”