Monday, May 25, 2009

Deflation or Disinflation?

Last week's April CPI report shows inflation continuing to shift downwards, mainly from a drop in food prices (log changes in CPI, 2005=100):

We're now back at Malaysia's long term average of 3% annual inflation, which on the face of it looks fine. Disinflation was always on the cards with the slow down in the economy, as well as the sharp drop in oil and commodity prices from their July 2008 peaks. With the rate of descent in economic indicators also slowing, deflation is less of a concern than it was in the early months of this year - or is it? The commodity bubble of 2007-2008 royally screws up any attempt at getting a feel for where inflation actually is:

Certainly the index level for April is showing the price level dropping as well, which is indicative of deflation (2005=100):

We're not just talking about the base effect here, folks. Month-on-month changes are also in the negative, except for the spike during Chinese New Year (log monthly changes in CPI, 2005=100):

I've tried constructing alternate measures, using 12-month, 6-month and 3-month moving averages:

Only the annual series is still pointing upwards as of April, which doesn't exactly fill me with confidence.

Worse still, the Jan-Mar 2009 manufacturing industry numbers show annual wage growth (-8.9%) dropping faster than employment (-7.0%), which suggests average wages dropped by 1.9%.

This is the critical channel by which deflation feeds on itself by reducing demand, which leads to further reduction in prices, which forces companies to lower output, and leads again to lower employment and wages. So although prices might be coming down, which is good for consumers, real income for those very same consumers is falling faster.

On that basis, I'm not convinced that the deflation threat is over yet. What comfort I have is that manufacturing is the worse hit sector in this downturn, and wage trends there might not be reflective of the economy as a whole. Also, anecdotal evidence from a reliable source (my mom!) suggests food prices have come up again in May, so we could see some upward movement in the index for this month. But I'm not holding my breath.


  1. Hi, I'm no econs/finance/accounts grad, but I have been following your blog posts for awhile already.

    Out of curiosity, how do you do your data mining and manipulation? Where do you get your data from and what software do you use to massage it + produce visual representations?

  2. Hi METALRAGE, thanks for dropping by.

    I'm preparing a post on data sources on the Malaysian economy, as I've gotten a lot of hits off Google asking the same question. It's a long post, so it might take a while before it gets published.

    The short answer is my data comes primarily from BNM's Monthly Statistical Bulletin which is available online, supplemented by data from the Department of Statistics and the Economic Planning Unit. I'll also use data from the IMF and World Bank as and when necessary.

    The software I use is EViews, which is a commercial statistical package designed specifically for econometrics. It's not the cheapest (I got it half price with a student discount), but it's relatively more user friendly than the others.

  3. Thanks Hishamh! Looking forward to your post.