The fault in our stats
Credit to Author: BEN KRITZ, TMT| Date: Sat, 08 Dec 2018 16:17:36 +0000
THE top news story on Friday was that both the business and consumer confidence index (CI) for the fourth quarter plummeted to levels not seen in years; the obvious conclusion is that these results indicate deep pessimism about the state and direction of the economy. But on closer examination, the conclusion might not be so obvious after all.
Business confidence and consumer confidence are two of the several dozen standard measures of the country’s economic health. The CI is the result of a survey — one for business firms, another for ordinary consumers — that basically asks whether the respondent feels the economy will improve, worsen, or stay the same from his (or the firm’s) point of view during the current quarter compared to the previous quarter. The score is the difference between the percentage of respondents who answered positively and those who gave a negative or neutral response. If the CI is positive, it indicates a majority of businesses or consumers are optimistic about the economy.
For the fourth quarter, the business CI registered 27.2, a positive result, but a drop of nearly 3 percent from the previous quarter and the lowest since the first quarter of 2010. Businesses are a little more upbeat about next quarter (Q1 2019), but only just; the quarter-ahead CI was 29.4, the lowest since the third quarter of 2009.
Consumers were far more pessimistic than businesses about both the current quarter and the one ahead. Consumer confidence for the fourth quarter slid deep into negative territory at -22.5 percent, and consumers’ outlook for the first quarter of next year remained gloomy, with the CI registering just -0.8 percent.
The “less optimistic” sentiments, as the BSP put it, of both businesses and consumers were attributed to the same things: Higher prices, higher interest rates, and a weakened peso. Businesses were also concerned about reduced sales volumes and lower supplies of raw materials, while the lack of wage growth and higher unemployment was said to weigh on consumer sentiments.
And all of that means…what, exactly? Although the CI is assiduously reported — in the United States, there is even an organization that does nothing but report it, and has even trademarked the phrase “consumer confidence index” — if you ask ten different economists or policymakers to explain in clear terms what the significance of it is, you are likely to receive ten different answers. The reason why is that there are two basic problems with the CI that make it virtually useless.
The first is one that Central Bank Deputy Governor Diwa Guinigundo highlighted in his comments about the fourth quarter CI: To give it some validity for the current quarter, the survey has to be conducted as close to the beginning of the quarter as possible; in this case, the two surveys were done during the first two weeks of October. As Guinigundo pointed out, none of the respondents have the benefit of foreknowledge of what may actually develop during the quarter; the survey period actually straddled the release of the September inflation figures, and of course at that point no one knew what the October or November result would be. Likewise, the sharp drop in oil prices and the moderate recovery of the peso were factors that almost certainly would have changed many responses.
The second problem is that the CI is not actionable. Although it is considered a leading indicator — in other words, business or consumer sentiment is supposed to change the trajectory of the economy moving forward — in actual practice it is not certain that it is. A good indication of this is the difference between the quarter-ahead business CI in the third quarter (i.e., the outlook for the current quarter), which was 42.7, versus the 27.2 current-quarter business CI. The outlook apparently changed based on what happened in the third quarter. Even though factors like inflation and exchange rates can be identified as affecting sentiment, because sentiment is what is being measured, there is no way to quantify the effects; different businesses and individuals will interpret the same factors in different ways.
That makes a working definition of business or consumer sentiment “an average qualitative judgment of the near-future state of the economy from a point in the recent past, based on subjective assessments of economic conditions prior to that point.” Figure out how that can actually be useful to economic policy or even provides a substantially accurate measure of economic health on a national scale, and they’d probably give you the Nobel Prize for it.
The bottom line is, while the CI does give statisticians something to keep them from browsing YouTube videos on company time, and while it makes for eye-catching headlines when it is a particularly ugly or attractive number, it’s not worth much more than that, and probably shouldn’t cause anyone undue excitement.
ben.kritz@manilatimes.net
The post The fault in our stats appeared first on The Manila Times Online.