What’s Behind The Drop In FDI?

You may have heard that inward FDI to Malaysia crashed in 2009 – Tony Pua has a nice roundup on his blog. You can read the full report from UNCTAD here.

As far as knowing the cause, I haven’t a clue…yet. There is of course the general drop in FDI across the globe in 2009 due to the global recession, but that doesn’t explain why Malaysia’s inward FDI dropped more than most. There’s also some interesting anomalies with the data that I really would want to know about before commenting on the situation, such as the massive increase (like, 100%+ a year) in both inward and outward FDI stocks from 2007-2009. This isn’t something that is confined to Malaysia – it runs across the global database.

As far as suspecting why inward FDI has been so poor lately, however, I have a few ideas.

August Industrial Production: Don't Mind The Dip

As before, and for the duration, I'm going to focus on levels rather than growth data. For anyone who's interested, here's the month-on-month growth for Industrial Production Index and its components (log monthly changes, seasonally adjusted):



Annual growth data of course isn't terribly informative with all the growth figures continuing to be in the red, save for the electricity index. If you've followed my last few posts, you'll know that even standard seasonal adjustment methodologies aren't a panacea, since fasting month follows a lunar not solar, calender. As such, August's slight downturn in seasonally adjusted month-on-month growth for all the components save manufacturing, shouldn't be taken too seriously and shouldn't be taken to be the start of something more pernicious.

Looking at the index levels shows a picture of gradual and moderate recovery in industrial output, save for mining, with July and August flattening out (2000=100):



This could in fact be taken to be a good sign, if my hypothesis that fasting month impeded output growth in August and September turn out to be true - more so since the manufacturing index has continued to push higher. On that basis I'd expect much stronger overall growth beginning this month, which we should see confirmed by mid-December.

The level of the electricity index - it's not far off it's all-time high - prompted me to think about capacity utilisation. We don't have official published surveys on this, so speculation is as far as it goes, but fitting trendlines (beginning 2000:01) to the indexes seemed like a reasonable rule of thumb approach to figuring out the gap between production and available capacity, so here goes:

Industrial Production Index


Mining Index


Manufacturing Index


Electricity Index


Adding together the gap between previous highs and the trendline, and the gap between current output and the trendline, as the current output gap, I got these numbers:

Overall IPI: 82.4%
Mining: 89.0%
Electricity: 90.6%
Manufacturing: 78.6%

The manufacturing number appears to too high to me based on media reports, but that's an artifact of assuming that peak production coincided with full capacity utilisation. So these numbers (obviously very rough) would actually represent an upper bound to current capacity utilisation. Still this may be useful barometer to follow in the future.

Technical Notes:
1. August Industrial Production Report is available from the Department of Statistics.

Deconstructing IPI Numbers

I finally got around to stitching together the IPI numbers - literally. As my last post on this subject mentioned, DOS has changed the basis of the index to 2005, which means some fancy footwork with splicing the different indexes together. The reason why I wanted to do this is to get a relatively longer view of trends within the industrial sector - which you can't do when the current 2005 index only runs for two years in the monthly series.

In any event some interesting things emerged. Here are the annual log changes in the main indexes, going back to 1999 (which was about the most I had patience to do today):



Note that mining seems to be holding up pretty well - pity it's only got a weight of 23.37. Manufacturing is of course the primary contributor to the downturn. Here are the actual indexes (2000=100):




In terms of output, we're approximately back to the level of 2004. If we take the maximum IPI level as representing full capacity utilisation (there's always some slack, so 90% would probably be a better approximation), then industry is running at about 80% capacity, and manufacturing alone at about 73%. That doesn't strike me as being sustainable for any length of time.

Of course, the main problem we have is with the electronics sub-sector (the electrical side of things has long been declining):



E&E output is now back to the same level as August 1999, nearly a decade ago. In terms of the drop in output, it's about 50% below the peak which was reached a short 15 months ago. Rubber, wood and chemicals also show sharp drops, if not quite as catastrophic as E&E:



In domestic manufacturing, the biggest drop is in construction related manufacturing:



All in all, not a pretty picture. The biggest concern is less the sharp declines in output, but rather how long the downturn will last. At this rate, with virtually no slowdown seen in output declines, there is going to be a lot of destruction on the ground of Malaysia's manufacturing capacity if this goes on for more than a few months. Which in turn implies that the risk in the banking system is significantly higher than most people expect.

The inclusion of RM25 billion loan guarantees in the 2nd stimulus package looks better by the minute, especially since it will be available almost immediately. I'm not sure it will do more than buy a little time though.