On Monday, The Manila Times and one or two other newspapers reported what seemed to be a bit of upbeat business news: The research arm of ratings giant Moody’s, in its Asia-Pacific Preview report, estimated that the Philippines’ manufacturing output in January increased by as much as 8.1 percent, driven by ‘buoyant’ global demand and continued low oil prices.
The forecast was part of Moody’s Analytics’ regular Asia-Pacific Preview, a roundup of upcoming economic indicator releases by governments in the region; the Philippine Statistical Authority (PSA) disclosed the January manufacturing figures on Tuesday.
The official release of the figures meant that by early Tuesday evening business editors all over town were scratching their heads wondering how they were going to rationalize the headlines they would have to publish on Wednesday with the ones they had already published Monday, because manufacturing in January did not quite live up to expectations. Instead of expanding by 8.1 percent, manufacturing output actually contracted in real terms; Moody’s forecast was not even close.
Factory output on a national scale is measured by a tool called the Monthly Integrated Survey of Selected Industries, or MISSI, which has five basic components: the volume of production index (VoPI) and the value of production index (VaPI); volume and value of net sales (VoNSI and VaNSI, respectively); and capacity utilization. Taken altogether, the MISSI provides a reasonably reliable assessment of the level and trajectory of overall economic activity. Although the indicators complement each other and should be looked at as a group, the most meaningful one is actually the value of production index, but the one that is usually cited by news reports and economic analysts when they make statements like, “Manufacturing in January likely accelerated to 8.1 percent” is the volume of production index.
The preliminary VoPI in January (preliminary, because the pattern is that the figure will be adjusted downward by a couple tenths of a percent in the next two months) was 3.3 percent, which, while still a positive figure, is not anywhere near 8.1 percent. The value of production index – in other words, what the country’s manufacturing output is actually worth in monetary terms – contracted by 1.8 percent. The net sales indices, which describe the movement of manufactured products out of factory inventories, were both positive; the 8.1 percent figure offered by Moody’s coincidentally matched the preliminary VoNSI figure, and value of net sales increased by 2.8 percent. Utilization of production capacity averaged 83.2 percent overall, with 13 of 20 industry groups averaging over 80 percent.
Signs of cooling
Before we get into why Moody’s got it so wrong – which is more a philosophical question than an empirical one – let’s break down what these figures actually say about the state of the economy. Because of the seasonal impact of the long Christmas holiday, comparing January’s factory output to that of the preceding December is pointless; year-on-year, the indications are that the economy, while still energetic, is slowly cooling. Capacity utilization was unchanged from a year earlier but factory output still increased, although at a slower rate than in January 2014. That’s generally a good sign, because it means manufacturers are producing at least a little more with the same amount of effort. The increase in volume of net sales is generally a good sign as well; it means that consumption remains robust. Both the volume of production and volume of net sales figures for January were supported to some degree by low oil prices, and may have also received a short-term boost from the visit of Pope Francis during the month.
On the other hand, lower prices probably had a significantly adverse effect on the ‘value of production’ index and dampened growth in the ‘value of net sales’ index. Thus, the overall situation is one in which factories are producing more with the same capacity, thanks to lower commodity prices, but the value of the output is gradually decreasing. Fortunately, growth in production volume is more than matched by growth in sales volume (again, thanks to low prices), but the increase in volume is not enough to completely compensate for the decline in value; VaNSI only expanded by 2.8 percent in January, while VoNSI increased by 8.1 percent.
The bottom line to all of the above is that while production and consumption are still robust, producers’ margins are slowly but surely declining. If that pattern continues through a quarter or two, it will start to reflect on manufacturing activity – production volumes and capacity utilization will drop – and that will in turn eventually result in decreasing consumer activity.
That does not necessarily mean the economy is in some kind of crisis; the data indicates that the economy may have peaked and is gradually slowing down, not that it is on the verge of collapse. It is in no sense, however, indicative of the continuing untrammeled growth Moody’s implicitly predicted. So how did they get it so wrong?
One lesson economists have had a hard time driving through our thick skulls is that the predictive power of economic models is in reality laughably weak. Trends and patterns only become clear and objectively valid over long periods of time; the shorter the prediction window, the more variables have to be taken into account, and the larger the margin of error becomes. Moody’s, it would seem, made an elementary error in trying to peg January’s anticipated factory output to a point on what they think is a longer trendline, which is a bit like trying to hit the Moon with a rifle shot.
That won’t stop anyone from making forecasts, of course; after all, analysts are analysts. Their predictions, however, ought to be regarded with a healthy level of skepticism.