The Message of the Fed Cleveland Median CPI
(April 23, 2006)
Dear Subscribers and Readers,
Note that a press release from us went out on Friday stating our position on the U.S. long bond.
We switched from a 25% short position to a neutral position in our DJIA Timing System on the morning of October 21st at DJIA 10,265 – giving us a gain of 351 points from our DJIA short on July 14th. On a 25% basis, this equates to a gain of 87.75 points. We switched to a 25% short position in our DJIA Timing System shortly after noon on Wednesday, January 18th at DJIA 10,840. We then switched to a 50% short position (our maximum allowable short position in order to control for volatility in our DJIA Timing System) on Thursday afternoon, January 19th at DJIA 10,900 – thus giving us an average entry of DJIA 10,870. As of the close on Friday (11,347.45), this position is 474.45 points in the red. We then added a further 25% short position the afternoon of February 27th at a DJIA print of 11,124 – thus bring our total short position in our DJIA Timing System at 75%. We subsequently decided to exit this last 25% short position on the morning of March 10th at a DJIA print of 11,035 – giving us a gain of 89 points. We subsequently entered an additional 25% short position in our DJIA Timing System on Monday morning (March 20th) at a DJIA print of 11,275. We had been trying to get rid of this latest 25% short position over the last couple of weeks. We had opted to wait for a more oversold condition before covering it – but it was not to be, as the Dow Industrials rallied a whopping 194.99 points last Tuesday (partly because of the March 28th Fed minutes released in the early afternoon). Again, we are now 75% short in our DJIA Timing System.
As of this point, we do not anticipate changing our positions in our DJIA Timing System anytime soon. In last weekend's commentary, we mentioned that many of our “ultra short-term technical indicators” were getting significantly oversold. Since then, they have become overbought once again. However, there are no potential triggers for a reversal in the near-time horizon – even as liquidity continues to decline and even as commodity prices are blowing off to the stratosphere. Should the market make a new cycle high right before the May 10th Fed meeting, there is a strong likelihood that the market will sell off right after the Fed meeting. Again, we will let our readers know on a real-time basis once we have decided to make changes to our positions in our DJIA Timing System. We will email you as well as post a message on our discussion forum (for folks who have set filters in their email software) letting you know (our message on our discussion forum will be titled: “A Change in our DJIA Timing System”).
So much for all the recent GM bashing – I bet many of our subscribers are getting sick of it now. So readers may find this a bit refreshing: For the year end 2005, the GM pension plan actually managed a return of approximately 13%, compared to a total (capital appreciation plus dividend yield) return of 4.9% in the S&P 500. According to Fortune Magazine, this translates to a return of $10.9 billion – or approximately the amount of money that the company lost last year in their operations. Starting in 2003, the company's pension plan became significantly more aggressive – by first borrowing $14 billion for the pension fund and then radically changing their investment policy by investing a “big chunk of new money [in] hedge funds and other alpha-chasing vehicles, such as emerging-market securities, private-equity firms, junk-bond portfolios, and derivatives.” Multiply this same strategy around the country's pension funds and individual 401(k) accounts and you get a world where international and emerging market inflows are at record highs and where commodity prices are blowing off. Okay, readers should now know that I am always a worrier at heart – so what happens once these markets top? Answer: There is no way a pension fund as big as GM's can get out in time.
Right after the release of the March 28th Fed meeting minutes, the market rallied as the odds (as determined by the Fed Funds futures) of a further rate hike in June to 5.25% declined from 54% to 30%. The metals did not disappoint either--as a Fed friendly environment is inherently bullish for cash flow negative assets such as commodities. While the metals are not as important as crude oil or natural gas when it comes to gauging potential inflationary pressures, they are definitely a reflection of the speculation that is currently occuring in the commodity markets. By far the most important commodities to watch, however, are (in this order) crude oil prices, natural gas prices, and steel prices. Readers should note that many steel companies will be reporting quarterly earnings this week, and you can bet that not only Wall Street analysts will be watching their projections, but that the Federal Reserve governors will be watching as well. Also, as the Fed had pointed out, there is a significant amount of anecdotal information suggesting that rising commodity prices are finally being passed on consumers, such as airlines hiking ticket prices because of high gasoline prices (effectively for the first time since 9/11), rising rental costs, and so forth. Given that the labor market remains very tight – and combined with decreasing productivity, the chances of a further rate hike beyond May 10th still remains high, in this author's opinion.
As I have mentioned before, it is this author's opinion that the Fed won't stop hiking (in the absence of an adverse aggregate supply shock) until 1) commodity prices decline substantially; 2) the Dow Industrials and the S&P 500 decline 10% to 15% from current levels, or 3) the current account deficit of the U.S. shrinks substantially.
So far, we have had none of this – but eventually, monetary policy will work – and both the stock and commodity markets will need to endure a substantial correction. Sure, the easy monetary policy of the Bank of Japan has clouded the situation a little bit – but even the Bank of Japan is now taking off their foot off the accelerator for the first time in a long time. Monetary policy has always worked – albeit with a 12 to 18-month lag. Remember how many commentators were remarking that the Fed was “pushing on a string” in 2002? Well, both the stock and commodity markets have never looked back since. Again, folks who are currently long the energies or the metals are now fighting against the Fed – and while the Fed was prepared to halt its series of rate hikes (as they had indicated in the March 28th minutes), it is difficult to imagine that the same Fed governors could even dare to stop the Fed's rate hike campaign today, given the huge outperformance of the energies and the metals subsequent to that meeting, as shown in the following daily chart:
Even natural gas – the laggard among the metals and energies – is up nearly 7% since the close on March 28th. Crude oil is up over 11%, and copper, zinc, and silver have just gone through the roof. More importantly, there are no good low-priced substitutes for any of these commodities. Sure, one can use plastic pipes in place of copper pipes, but given that crude oil and natural gas prices are still rising, it is difficult to see how the price of plastics cannot go up as well. Unless the U.S. can ramp up her nuclear power plant construction or commercialize the production of carbon nanotubes in the next couple of months, this author believes that the chance of more Fed rate hikes beyond the May 10th meeting is high.
Please note that many of the “anecdotal inflation pass-through” are now being captured by the CPI data as well – as the latest CPI-U data for March from the Bureau of Labor Statistics surprised on the upside (mostly due to “shelter” and “transportation”). More importantly, the latest “median CPI” data released by the Federal Reserve Bank of Cleveland also surprised on the upside, as the annualized latest month-to-month change in the Cleveland Median CPI hit 5.0% - the highest in over 11 years.
So Henry, what is the Cleveland Median CPI? And why is it important?
Let's first try to answer that first question. As the title suggests, the Cleveland Median CPI is an inflation indicator which attempts to measure the price change of the middle observation of the goods and services that are contained in the CPI-U basket. Quoting from the Cleveland Fed: “In effect, the median consumer price change is the CPI less everything but the price change that lies in the middle of the continuum. Since only the order, not the values, of the various price changes is used in its calculation, the median is a central tendency statistic that is largely independent of the data's distribution. The median also has the intuitively appealing property of lying close to the majority of price changes than does any alternative measure.”
The Cleveland Fed claims measurements such as the Consumer Price Index measures “only” the “average price of an array of goods and services purchased by households, but because it is constructed as a weighted mean of all consumer prices, it does not discriminate between relative price changes and inflation. Indeed, the CPI may rise when the price of just one commodity increases.” The Cleveland Fed claims that the traditional CPI may not be a good measurement tool of inflation, as “an increase in one price relative to others is the signal that directs resources and rations consumption [in other words, this does not take into account the “substitution effect” by consumers that can ultimately depress prices of the good that has gone up].” Meanwhile, inflation is “a monetary phenomenon that determines the underlying level of all price changes; it has virtually nothing to do with the transmission of market information. Indeed, one fundamental problem with inflation is that it can be confused with relative price movements, obscuring the transmission of market information and reducing market efficiency.”
That is, the Cleveland Fed claims that certain non-monetary events can “at least temporarily, distort reported inflation statistics … during periods of bad weather, for example, food prices may rise to reflect decreased supply, producing transitory increases in the CPI. But since these prices do not constitute monetary inflation, monetary policymakers may want to avoid including them in their decision-making.” In his book, “Inflation Targeting,” current Fed Chairman Ben Bernanke outlines precisely such a scenario (one that will involve the Fed stopping or even easing as opposed to hiking). I will repeat the relevant quote from our last commentary: “In general, there is no conflict between output and inflation stabilization when the precipitating shock is an unexpected change in aggregate spending; using monetary policy to offset an aggregate demand shock is nearly always the correct response. However, an aggregate supply shock, such as a sharp increase in oil prices, may cause a conflict between stablizing output and employment in the short run and stablizing inflation in the long run. Targeting a price index that excludes the first-round effects of common supply shocks can, as we have seen, ameliorate this conflict to some degree. But a supply shock that is great enough, or that arises from some unanticipated source, may justify missing or changing a previously announced inflation target.” The Cleveland Fed claims that such an evaluation can be done more precisely by using the median CPI, as opposed to the currently-popular method of the “core CPI” – which is basically the CPI excluding food and energy (together, they constitute more than 25% of the CPI). Says the Cleveland Fed: “One commonly used technique for measuring underlying or core inflation is to exclude certain prices in the computation of the index, based on the assumption that these prices are the ones with "high-noise" components. This is the rationale behind the commonly reported CPI excluding food and energy data. However, economists Michael Bryan and Stephen Cecchetti have found a measure that forecasts inflation better than either the CPI excluding food and energy or the all items CPI: a weighted median of the CPI.”
Put more simply, the Cleveland Fed claims that Bernanke can do a better job of evaluating such “supply shocks” by looking at the readings of the Cleveland Median CPI than solely relying on anecdotal information. Now that we have settled this issue, what is the Cleveland Median CPI saying now? In brief, the latest month-to-month change reading does not look good – as the annualized monthly increase in the Cleveland CPI hit an 11-year high of 5%. Following is a monthly chart showing the month-to-month changes in the Cleveland Median CPI (annualized) vs. the effective Fed Funds rate from January 1990 to March 2006:
On inspecting the above chart, one can certainly make the claim that the Fed had already paused and even started easing the last time the Cleveland Median CPI hit multi-month highs (such as during January 1991 or during 2001). However, readers should note that the spikes during 1990/1991 and 2000/2001 all occurred while the economy were slowing down and drifting into recession (which were associated with both a declining stock market and a declining commodity market). Today, the economy is still going strong – with the latest ECRI Weekly Leading Index readings (the annual rate-of-change is now at 3.0%) turning up and with both the Dow Industrials and the commodity markets hitting multi-month (and even multi-year) highs. Moreover, the new Bernanke Fed has come out and stated their position on focusing on an “inflation target” as opposed to the Greenspan Fed – who focused on both maintaining strong economic growth and maintaining a reasonable inflation number. Certainly the Fed cannot merely stop with another quarter-point rate hike by May 10th – unless the stock and commodity markets crash between now and the June 28th meeting (the next Fed meeting after the May 10th meeting).
The wild card, of course, is U.S. housing, and while there are already signs of a housing market slowdown in parts of the country (combined with a slowdown in refinancings and mortgage equity withdrawals), we haven't seen an across-the-board slowdown at this point – even as the yield of 30-year Treasuries has risen substantially in the last six months. The lack of an across-the-board slowdown thus far is being reflected in the lack of a significant correction in our MarketThoughts U.S. HomeBuilders' Index – which is essentially an equal-weighted index comprising of the five largest homebuilders in the U.S. (KBH, PHM, LEN, CTX, and DHI):
Could the housing market and the homebuilders slow down quickly enough before June 28th for the Fed to justify at least a pause in its hiking campaign? Two months can be a long time in the financial markets – but given what is happening in the Cleveland Fed Median CPI, and the commodity and stock markets at this point, my belief is that we will need at least one more rate hike after the May 10th meeting (or a 50 basis raise during the May 10th meeting) before the Fed should pause or stop its rate hike campaign.
Let's now turn our focus on the stock market. While the Utilities and other stocks (such as poultry, Google, and Apple) have bounced back in the last week or so, breadth in the U.S. stock market is still relatively weak, as outlined by the following three-year chart showing the Wilshire 5000 vs. the number of new highs vs. new lows in all stock market issues, courtesy of Decisionpoint.com:
As shown by the above chart, the number of new highs vs. the number of new lows in the U.S. stock market has been on a general decline since peaking in January 2004. This is also being reflected in other breadth indicators, such as the McClellan Summation Index and the percentage of stocks above their 200-EMAs. More recently, the indices and sectors that have historically led the market have been very weak, such as the Dow Utilities (which peaked in early October 2005), and the Consumer Discretionary sector. This makes sense – as consumer discretionary spending (flat-screen TVs, vacation trips, Coach Briefcases, etc.) is usually the first to take a hit in an economic slowdown. Unlike the last cycle (when the recession was primarily driven by a plunge in capital spending), this current cycle has been driven primarily by the housing sector (mortgage equity withdrawal, creation of jobs in the housing sector, etc.) and the resultant increase in consumer spending because of the booming housing market. Therefore, any slowdown in the economy or the stock market going forward should first been reflected in the stock prices of the consumer discretionary sector. Again, using consumer discretionary stocks as a leading indicator is all the more important in this cyclical bull market, since the recovery from the 2001 recession has very much been consumer-driven. Below is a weekly candlestick chart showing the AMEX Consumer Discretionary Select Sector SPDR (XLY) and its relative strength vs. the S&P 500 from January 2002 to the present (courtesy of Decisionpoint.com):
As mentioned on the above chart, the absolute level of the XLY peaked in January 2005 and has been trending down and making lower highs ever since. More importantly, the relative strength of the XLY vs. the S&P 500 also peaked in January 2005 and is now in fact at the lowest level since March 2003. Weekly stochastics have also been at overbought levels and are now trending down. Historically, the consumer discretionary sector is an “early cyclical” and thus is a very good leading indicator of the stock market, especially since this cyclical bull market has mostly been consumer-driven. The top ten holdings (in order of percentage makeup) of the XLY are Home Depot, Time Warner, Comcast, Lowes, Viacom, eBay, Target, Disney, McDonalds, and News Corporation. Other notable holdings include Carnival, Starbucks, and Best Buy. Again, this author is now watching Starbucks and Best Buy like a hawk, as these two stocks have been two of the hottest stocks (with still very compelling growth stories) in this cyclical bull market and is still near or at all-time highs. Once either Starbucks or Best Buy takes a hit, this cyclical bull market in all likelihood will be over. Note that both Starbucks and Best Buy are still trading at near all-time highs and are not even close to breaking down yet at this point. The questions are: How much longer will folks pay up for $3 coffee (after all, this cheapskate of an author is still doing it)? And will the introduction of Blue-ray DVDs and HD-DVD spur another round of consumer buying later this year? We will just have to see.
Let's now discuss the most recent action in the stock market. The market certainly surprised many folks on the upside last week, as the Dow Industrials rose by 209 points while the Dow Transports rose by 60 points. As I pointed out in our last weekend's commentary, however, many of our ultra short-term indicators were in oversold territory, and thus the huge rally on Tuesday should not have come as a total surprise to many of our readers. For now, the market is most probably in a consolidation phase – and while it is difficult to tell what the short-term trend may be, there is no doubt in this author's mind that from now until the end of this year, the intermediate trend is definitely down. Following is our daily chart showing the action of the Dow Industrials vs. the Dow Transports from July 1, 2003 to the present:
In the meantime, the Dow Utilities embarked on a major bounce last week, as the index rose by 16.87 points to close at 398.36 last Friday. Please note, however, that this is still many points away from the all-time bull market high of 437.63 made on October 3, 2005 – and given that bond prices have shown no signs of life over the last week or so, it is doubtful that the Dow Utilities could continue to build on last week's gains in the weeks ahead.
I will now end this commentary with a quick discussion of our popular sentiment indicators – those being the bulls-bears percentages of the American Association of Individual Investors (AAII), the Investors Intelligence, and the Market Vane's Bullish Consensus Surveys. Readers who want us to periodically show you the individual weekly charts should let me know – but in the meantime, I will only report to you a condensed format of our three popular sentiment indicators. The message continues to remain the same as last week's: While the short-term readings of these indicators may be more on the neutral side (relative to the readings we have gotten over the last couple of years), readers should note that these readings are still pretty overbought relative to the readings since January 1997.
During the latest week, the four-week moving average of the bulls-bears% differentials of these three popular sentiment indicators declined to slightly oversold levels from 24.7% to 23.3%. Following is a weekly chart showing the four-week moving average of the Market Vane, AAII, and the Investors Intelligence Survey Bulls-Bears% Differentials from January 1997 to the present week (note that I have redid the scales and shortened the time period by changing the starting month from July 1987 to January 1997):
The fact that this sentiment indicator is still tilting towards the neutral side tells us that there is still more room to run on the upside – but given the weakening breadth numbers, tightening liquidity, and the overbought conditions in our other technical indicators, this author won't be willing to bet on the long side here until our sentiment indicators get significantly move oversold (for example, readings that would put us at the October 2005 levels.
Conclusion: In light of all the evidence we have presented in this commentary and in many of our commentaries over the last several weeks, the chances of the Fed stopping at 5.0% at the May 10th meeting is not a foregone conclusion – and in fact, the odds of a 5.25% Fed Funds rate in the months ahead is rising by the day. Unless the stock, commodity, or the housing markets slow down dramatically in the coming weeks, the Fed is basically compelled to at least hike two more times – even though we all know monetary policy usually has a 12 to 18-month lag.
In the short-run, this author will continue to remain 75% short in our DJIA Timing System – with no plans to change our signals during this upcoming week. For now, stock selection (both on the long and the short side) continues to be the key. As for bonds, the short-term and intermediate-term trend is still down, although we will continue to keep an eye on bonds and will most likely do some kind of trade on the long side within the next one to two months.
Henry K. To, CFA