Thursday, November 19, 2009

Deflation Watch (October 2009): Price Level Trends Relative to Past Debt Crises

This is another update (with October's data) to the series of posts on US price level trends that started with Price Deflation Today versus the Great Depression and Post-1990 Japan — Comparative Charts (which had data through July). The original post contains the most in depth discussion of the comparisons between the three episodes, so please look at that if you have not already.

I'm posting updated charts as the data is released each month, but for now since time is scarce I will keep the commentary minimal... I will have more to say if and when the current price level trends change more significantly.

October shows a minor increase in price level for many components of the CPI. According to the BLS much of this is energy and auto related, but while disinflation is still apparent when looking at 2009's trend, strong deflation has yet to reappear in any CPI components. The big question is if and when this will change. While I expect it will, time will tell.

These are my definition related comments from a previous post:
As noted previously, "deflation" is often discussed in broader terms than simply price level:
  • Contraction of money and credit (broad money supply)
  • Deflation in asset prices
  • Deflation in a representative "basket" of consumer and producer prices
  • Deflation in wages
The various measures are often somewhat correlated but they only track to each other loosely. In the Great Depression prices fell faster than wages, yet wages (along with asset prices) still fell enough to propagate the adverse feedback loop of debt deflation in which income falls but debt obligations remain at the same nominal level, increasing the burden of the debt. Deflation in asset prices (triggered by the bursting of debt-financed asset bubbles) generally precedes the other deflationary trends.
Some Relevant Current Articles
  • Apartment Rents "Plunge" in the West — The BLS rent measure still seems to lag reality.
  • More on Falling Rents — The rental vacancy rate is at record levels, pressuring rents and ultimately home prices lower.
  • Bill Gross: Fed on hold through 2010Pimco's Bill Gross apparently expects that unless we can sustain 5-6% nominal GDP growth over the coming years (a rate the economy is geared for) then debt deflation is likely to take hold. Unless I am misinterpreting these remarks, this is ominous, as true debt deflation dynamics would likely lead to an economy no one could deny was in depression.
  • What Stinkin' Inflation? PPI Edition — Breaks down the PPI changes in a useful graph. Food and energy increases dominated in October.
As I've alluded to in past posts, commodity prices have seemed far ahead of fundamentals. What's unclear is whether they will keep rising anyway, plunge again like in 2008, grind lower more slowly, or stagnate for years until fundamentals catch up. A crash or decline could feed through violently to CPI changes as in 2008, potentially amplifying other deflationary forces. A sampling of opinions include:
  • Nouriel Roubini, One on One: More Doom and Gloom"Well in commodities, I look at oil prices. They fell from $145 last summer, came down to $30 earlier this year and now they’re back close to $80. But if I look at the fundamentals of demand and supply, demand is down to 2005 levels, supply and inventories are at all-time highs. In my view, the movement in oil prices is not fully justified by the fundamentals."
  • Pig Farmers are Making Brent Nervous — on "pervasive" hoarding of metals by private speculators throughout China.
  • Goldman’s Global Oil Scam Passes the 50 Madoff Mark — On the ability of financial speculation in futures to drive commodity prices (the focus here was on oil).
  • Commodity inflation — James Hamilton asks "Why are the prices of so many commodities rising in an economy that seems to remain quite weak?"

Price Level Charts for October

CPI-U 12 Month Changes, 1999 to Present (US) (source: BLS)

(click on chart for a larger version in a new window)

BLS Summary Comments:
"On a seasonally adjusted basis, the Consumer Price Index for All Urban Consumers (CPI-U) rose 0.3 percent in October, the U.S. Bureau of Labor Statistics reported today. The index has decreased 0.2 percent over the last 12 months on a not seasonally adjusted basis.

The seasonally adjusted all items increase largely reflected advances in the indexes for energy and for new and used motor vehicles. The energy index rose for the fifth time in the last six months, advancing 1.5 percent as the indexes for gasoline, fuel oil, natural gas, and electricity all increased. The index for all items less food and energy rose 0.2 percent in October, the same increase as in September. The indexes for used cars and trucks and for new vehicles both rose sharply and together they accounted for over 90 percent of the increase in the index for all items less food and energy. The indexes for airline fares and medical care also increased, while the shelter index was unchanged and the indexes for apparel and recreation declined.

The food index also increased in October, rising 0.1 percent after declining in two of the previous three months. The index for food away from home increased slightly, while the food at home index was unchanged. Within the food at home group, the index for dairy and related products rose significantly, while the fruits and vegetables index declined for the fourth straight month."
16% Trimmed CPI

This chart of 16% trimmed mean CPI (generated from the Cleveland Fed site) removes the most extreme monthly price changes:

(click on chart for a larger version in a new window)

Consumer Price Index Trends: Great Depression versus Today through October 2009 (US)
(click on chart for a larger version in a new window)

Components of US Consumer Price Index (May 1927 - Dec 1937, Great Depression)

(click on chart for a larger version in a new window)

Components of US Consumer Price Index (Jan 2006 - Oct 2009)
(click on chart for a larger version in a new window)

Annualized 3-Month Rate of Change for Components of US Consumer Price Index (Apr 2006 - Oct 2009)
(click on chart for a larger version in a new window)

The above chart shows the rate of change (over a sliding three month period) of the components whose absolute price levels are shown in the previous chart. I also added the magenta line for shelter (even though it is contained within the yellow housing measure) to better show the effect of declining rents and owners' equivalent rents separated from other housing components such as energy.

Price Index Changes: Great Depression CPI versus Current PPI through October 2009 (US)
(click on chart for a larger version in a new window)

I added "Finished goods less food and energy" starting this month. It shows marked disinflation (and deflation for October, but that could be a blip for now).

Consumer Price Index Trends: 1990s Japan versus US Today (through October 2009) and US Great Depression
(click on chart for a larger version in a new window)

CPI in Japan (Jan 1980 - Jul 2009)

The peak of Japan's CPI occurred in October 1998, almost eight years after the stock market peaked, and Japan's notorious mild deflation has been in effect since then. A multi-year disinflation (of core CPI) leading to sustained mild deflation is one possible outcome for the US.
(click on chart for a larger version in a new window)

Factors Contributing to Deflation

The measures I've been showing here recently haven't changed significantly this month, so I'm going to skip updating them this time. I'll bring them back when there are noteworthy directional changes.

Saturday, November 7, 2009

Global Price Level Trends Show Varied Inflation/Deflation Across Housing Bubble Countries

While looking at US consumer price level trends in prior posts (starting in September), I have pondered the differences between the Great Depression experience in the US of severe deflation and Japan's post-1990 slower-onset milder deflation to assess today's risks. Factors that could in some combination explain this historical difference in severity include:
  1. A greater amount of government stimulus applied by Japan to maintain aggregate demand.
  2. Japan's post-asset-bubble balance sheet recession occurring in the context of a booming global economy, which would support demand for Japan's output via strong exports to the rest of the world, as well as put a floor under commodity prices.
  3. Consumer prices in modern economies potentially much "stickier" than was the case in the 1930s, due to a larger share of complex goods in our typical consumption "basket", plus a large services sector, neither of which seem likely to be dragged down rapidly by falling commodity prices. (Labor prices are are generally not renegotiated daily like commodity prices). If accurate, this could suggest modern economies are less likely to enter the vicious spiral of self-reinforcing deflation.
  4. Larger upward inflationary momentum at the time of the asset bubble burst. In the US, when the stock market and credit bubbles started bursting in 1929, three out of four CPI components were already in mild to moderate deflation. Japan had roughly 3% inflation at the time of the stock market bubble peak and it was almost eight years before the consumer price level peaked.
In gauging the accuracy of the above potential factors and their relevancy to today's crisis, it may be helpful to look at the price level trends of various modern economies that have experienced asset prices bubbles leading up to the 2007-2009 global financial crisis. This post will include graphs for the following countries:
  • the US (for obvious reasons)
  • the UK, Ireland, and Spain (as the three most bubbly major European economies, at least with respect to housing prices)
  • Japan (the world's second largest economy, and of interest to see what happens decades after asset bubbles burst)
  • Australia (as a major commodity exporter that also has a housing bubble, and apparently the only major economy to so far escape a technical recession)
  • the Euro 27 area (the largest aggregation of Euro countries offered by eurostat)
  • Iceland (a bit of an outlier in part due to massive externally denominated debts and different crisis dynamics to date, but of interest for comparison purposes)
I initially attempted to include China's data, but only year over year changes are provided for each month, so there is insufficient data to reverse-engineer back to the raw price level data.

Here is a chart of consumer price level trends for these countries, with all price levels normalized to 100% at July 2008 (which was the peak of the non-seasonally-adjusted US price level):
(click on chart for a larger version in a new window)

Some observations:
  • The US CPI has been the most volatile — the fastest rate of increase leading up to summer 2008 (excepting Iceland), and the most rapid deflation in the second half of 2008. Perhaps this reflects a greater CPI weighting to energy than the other countries.
  • Iceland's price level inflation has been so dramatic relative to other countries that rather than show the full climb (from 87% to 118%), I zoomed in the chart to make the differences between other countries more visible. I'm not sure to what extent this inflation was driven by falling currency exchange rate versus other factors.
  • Spain and Ireland (but especially Ireland) appear to be in the strongest deflationary trend among European countries included here.
  • Spain and Japan have tracked closely to Ireland since mid 2008, but are in 2009 showing a little more price level stabilization than Ireland.
  • The UK and Australia each paused in their price level ascent in the second half of 2008, but their prices have marched higher again in 2009. Probably not coincidentally, both countries have seen housing prices rebound and set new highs (I have read this but not confirmed the data).
The next two charts show the annualized trailing three month inflation rate for each country. The charts are identical other than the Y-axis scale — the second shows the data in full, while the first is zoomed in, excluding most of Iceland's data points, to better see differences in the other inflation rates. Note these are probably extra volatile because many of the data series are not seasonally adjusted.

(click on chart for a larger version in a new window)

(click on chart for a larger version in a new window)

Some observations:
  • All the countries except Iceland experienced a dramatic decline in inflation at the end of 2008. Even the least affected, Australia, still had a negative three month annualized inflation rate lasting three months. This is very likely the result of global commodity price declines in the second half of 2008, underscoring that some price level influences truly are global.
  • Iceland's inflation rate soared in early 2008 and has been declining ever since. It is still the highest rate of the countries I've shown, at 8% (July-September 2009, annualized), but has been dis-inflating rapidly from a peak three month annualized rate of around 30%. I have wondered whether at some point Iceland could fall into deflation, since at least a couple years ago its domestic credit levels looked quite high relative to GDP. I don't know what the domestic debt dynamics are currently and perhaps the dynamics of a huge external debt level contribute more substantially to macroeconomic developments.
  • Only Australia has had a three month annualized inflation rate that has climbed consistently through 2009 (September is the latest data). The UK diverges from Australia with its three month inflation rate declining since April 2009 (but it did increase faster at the start of 2009).
  • Excluding Iceland, the most recent data point, September 2009, shows a range from around 4% inflation (three month annualized) in Australia down to around -4% inflation (i.e., deflation) in Ireland.
Concluding Thoughts

This was a very rough high-level look at relative price level trends and short term inflation rates over a sampling of housing bubble countries and other countries of interest (Japan, Iceland). More insights would be likely with additional analysis such as adjusting for exchange rate trends, comparing rates of government deficit spending, comparing changes in private debt levels, etc.

The price level trends across countries show some correlation at times (e.g., late 2008), suggesting global macroeconomic factors matter to some degree, but also diverge significantly at other times (e.g., the current -4% to 8% range of 3-month annualized inflation across countries), suggesting domestic country-specific dynamics are the dominant factors in price level trends, even at times of global recession/crisis.

There are at least three key domestic dynamics that could be most heavily influencing price levels. One is the degree of government stimulus relative to the contraction in private spending. A second is whether private debt bubbles have actually popped or are still growing. A third would be the relative size of these debt bubbles (total debt-to-GDP) and their recent rate of growth (since added debt contributes to annual aggregate demand).

I have not included enough data in this post to evaluate these factors for each country. But at the risk of being wrong, my current assumption is that size of government response is the number one differentiating factor, and some countries are maintaining a stronger private-sector bubble mentality than others (e.g., with respect to Australia's housing prices). If government stimulus proves politically unsustainable, or private sector debt bubbles collapse and prove to be overpowering, it could be that countries like Australia will simply lag countries like Ireland with respect to consumer price deflation. Ireland may be the country to watch — I've seen suggested that it may have begun a full scale debt deflation.


Note that different countries construct their CPI measures differently, so some trend differences likely reflect what each country considers a representative "basket" of goods and services. Also the US, UK, Ireland, Spain, Iceland, and Euro 27 price levels are not seasonally adjusted. I think the data I found for Japan's and Australia's are seasonally adjusted, but I am not certain.

The reason I like looking at absolute price level trends right now rather than just rate of change (typically year-over-year) is that when price levels are not moving in a single direction (until the recent deflationary trends, the direction of CPI indexes was primarily up), then the year over year change only incorporates two data points at a time, and is therefore especially volatile and doesn't reveal the big picture as well.