Economic Outlook for 2006
Review of 2005
My Outlook for 2005 did not offer much promise of reward and it did not offer much concern about risk. Generally speaking, I felt that the indicators for the US market were mixed-to-neutral and I didn’t see much upside or downside. In fact, through August the market was up less than just a few percent.
Sure, there were intermittent periods of market pullbacks. For example, in the September 2005 issue of the Navigator Newsletter (just as the S&P 500 was touching new intermediate highs) I warned that the large cap index would dip to about 1,182 points by sometime in October (or about 5% from its recent high). We ended being off by six points (the index hit 1,176), but the point is that 2005 was a year of very narrow trading channels.
We thought that in the beginning of the year that this type of narrow trading range would happen, so we used a lot of foreign stocks and took a largely neutral position on US stocks (using dividend payers and defensive stocks like those in the Health Care sector). The good news is that we took the correct action to take advantage of the very few opportunities that were available last year. The bad news is that this year, in 2006, the market indicators are still mixed-to-neutral, but there will be fewer opportunities to take advantage of.
[back to top]
Summary
Central to the outlook for 2006 is the Fed’s tightening cycle. It may end with this year still young, with the yield curve inverted, and with rates broadly higher. If core inflation is also heading upward along with mounting signs of trouble in the housing market, then the chances will increase for a slowdown in economic growth and corporate profits as the year progresses. The chances would also increase for the stock market to weaken, consistent with the second year of an election cycle. The market’s pattern for that cycle reaches the year’s low point in the fall, suggesting that the S&P 500 would move closer to the bottom of our 1,023 – 1,404 risk/reward range.
Recognizing the heightened risks in a cyclical bull that’s now in its fourth year, we’re entering 2006 positioned defensively, currently favoring (albeit only slightly) larger -caps over small-caps, and Value over Growth, and perhaps soon to be favoring bonds over stocks (this would be especially true if the Fed stops its tightening cycle, bond yields peak at the upper end of their range, and economic growth in fact recedes). And we’re expecting out-performance from the defensive Health Care and Consumers Staples sectors, with under-performance from cyclical sectors. A second-half market bottom could, however, present an opportunity for recoveries in Growth stocks and small-caps.
Whatever scenario unfolds, we will be watching to see, as if in 2005, many of the best money-making opportunities remain outside of the U.S. financial markets. Entering 2006, the China-driven commodity demand theme remains well intact, underpinning the advance in commodities, emerging markets, and resource stocks globally. Emerging markets would be likely to under-perform in a global market downturn, and corrections in the other long-terms are also inevitable. But a severe global economic contraction could be needed to threaten their long-term sustainability.
More important than my current expectations, it’s my models and rankings that will keep us objective and flexible in 2006, guiding our strategies over the course of the year.
[back to top]
Forecasts
I don’t believe that anyone, including myself, can forecast consistently and reliably, but occasionally one can get lucky and get a forecast correct (or, in Berkshire’s blessed and fortunate case, be on a streak since inception). For the election year of 2004 I predicted a gain of above 8%, which is the mean and the median for election years since 1889. For 2004 the Dow Jones Industrial Average (DJIA) was up just 3.1%, but the S&P was a near bulls-eye at 9%. Since that worked for 2004 and because the weight of the indicator evidence was pretty mixed coming into 2005, I tried the Presidential Cycle again, which shows a mean gain of 2.9% (and a median gain of 7.1%) for the post-election year. The Dow under-performed and was flat, but the S&P 500 was a near bulls-eye as it ended right at 3%.
The indicators today suggest another single-digit gain for 2006, which would be in line with the historical mean and median mid-term election year returns (3.3% and 1.1%, respectively). The trouble with that forecast is that consecutive single-digit return years like 2004 and 2005 have only occurred twice over the last 117 years. And three consecutive years with single digit gains has never happened; I don’t want to catch myself saying “this time it is different.”
While the second Presidential year has tended to be pretty neutral (consistent with my indicators), it has also ended with a strong rally typically starting after a poor first half or weakness that may extend into October.
Regarding the impact of Fed tightenings (and whether or not this time it may be different), bad performances have not always followed after the Fed was finished raising rates. In particular, the markets did very well in 1989 and 1995. What seems to distinguish the good cases from the bad cases is the yield curve. It did not actually invert in 1989 and 1995, but it did in most of the bad cases. So the shape of the yield curve, and whether it actually inverts, is one of the key things that I am watching.
Last year I cited 1,300 points on the S&P 500 as the potential reward and 1,000 as the potential risk. We spent much of that year in the middle of that range. For 2006 I am focused on a number of valuation metrics. At extremes, my models would suggest over-valuation at 1,404 and under-valuation at 1,023. I would guess that most of the action next year would be in that zone and actually more to the middle. A key to keeping us between fair value (whatever that is) and overvalued will be continuing high profit margins. Normally, profit margins tend to fall from these current levels and thus the market does not gain much, consistent with another neutral year. However, some argue that productivity will keep profits strong, and I am counting on that.
Besides watching profits and interest rates (which continue to offset each other), the other key factors could knock down the theory of three (three consecutive years with single-digit market returns) would be a collapse in housing, if affordability continues to plunge, or a rise in core inflation to well above 2%.
In conclusion, the statistical odds against three straight single-digit gain years in a row are astronomical, so I am on a serious watch as to where I could go wrong. But for now, the evidence is indeed mostly neutral to slightly defensive.
[back to top]
Bonds and the Economy
For 2006, we have estimated economic growth from two “top-down”, or macro-economic, perspectives. Using projected ranges for some indicators, our first top-down forecast provides an estimate using the full date range of each indicator. The second top-down forecast uses just the past ten-years of data. Although the results for the individual indicators are quite different in some cases, the overall forecasts are remarkably similar.
The average of the two forecasts is about 3.5%, slightly above the Wall Street consensus of 3.3%, although I think that risk is to the downside, closer to 2.5 – 3.0%. We’re still looking for business spending to outpace consumer spending. Businesses spending should still benefit from transportation and infrastructure expenditures, while consumer spending should be hampered by a reduced amount of home equity extraction. Government spending should remain below trend growth (consistent with the second year of the election cycle, especially for a second-term President). And the trade deficit may act as a bigger drag on GDP growth later in the year.
Outside of terrorism, bird flu, and more hurricanes, the biggest risk to our forecast is excessive Fed tightening. Another risk is a collapse in housing demand and home prices, reducing equity extraction and eliminating most refinancing activity.
For 2006 we have estimated the CPI from both a top-down (i.e. based primarily on macro-economic growth) and a bottom-up (i.e. looking at prices for specific products/services). The top-down forecast is 3.6%, but I think that is high and that our bottom-up forecast of 2.3% is more realistic (the Wall Street consensus is 2.5%). More stable energy prices should slow the overall inflation rates. The core rate (CPI excluding food and energy is the core rate) may drift higher in the first half of the year before falling back to 2.0% by yearend.
There are both upside and downside risks to my CPI forecast. A continued rise in energy prices and a resumption in of the dollar downtrend are the biggest risks to the upside. Faltering economic growth from higher interest rates and a weakening housing sectors present risks to the downside.
For 2006 I anticipate that the 10-year Treasury note will remain within a range of 3-3/4% to 5-1/8%. Currently, our valuation model says that the 10-year notes are fairly valued around 4-7/8% (current the 10-year is around 4.50%). With the economy at risk of slowing and with inflation expected to remain behaved, fair value could fall below 4-1/2%. Additionally, good resistance can be found around 3-3/4% to 3-7/8%. Should yields approach fair value we may become more attracted to bonds as an investment.
Regarding the yield curve flattening trend (the yield curve is flat when two-year notes yield the same as ten-year notes, when typically ten-years yield more), it has lost some momentum recently. At most I expect a modest inversion in the first half of this year. Historically, flattening trends have ended around the last rate hike.
[back to top]
Stock Market & Style Allocation
For additional perspective on cyclical influences, we can combine the projected pattern of the four-your election cycle with the one-year seasonal cycle and the ten-year cycle to come up with a composite. In 2005, the DJIA’s trend was more subdued than the composite had predicted, though the approach proved to be more useful in predicting the trends of the NASDAQ and the S&P 500. In contrast to 2005, the ten-year cycle’s influence is not favorable for 2006. Whereas the fifth year has tended to be the best of the decade, with the S&P 500 gaining an average of 24%, the sixth year has tended to be ordinary, as the index has gained an average of 7%.
More important than the market’s cyclical status will be the status of market breadth, which has remained subdued and divergent over the latest rally. Several indices have refused to confirm the latest intermediate-highs in the S&P 500, and only three of the 10 S&P sectors reached highs during the rally. This bad-breadth (breadth being a measurement of stock price direction) is not a favorable indicator for the market.
As the market moves into 2006, optimism is relatively high, as exemplified by the average of 69% stock allocation by Wall Street strategists. The S&P 500 has declined at a 12% per annum rate when this indicator has been in this optimistic mode. If the hostile monetary environment turned friendly or if the unimpressive breadth turned decisively bullish – thus proving that the Wall Street strategists should be as optimistic as they are – then the chances of a cyclical bear market (i.e. a meaningful and lasting correction) would diminish. That is not an impossible scenario, as it happened in late 1994 / early 1995. But if conditions were to deteriorate it would be more like late 1976 / early 1977 when the DJIA failed to reach new highs and stock prices turned decisively lower as interest rates headed higher.
During the 1994-1995 period, the record-high DJIA corrected and entered a trading range when rates started rising. But when the rate pressures receded again, the DJIA returned to record highs with a break-out of good breadth. That development was lacking in the 1976-1977 period. So will 2006 bring a break-out of new highs (i.e. good breadth), accompanied by a development that would support an allocation shift upward? If not, a continuing lack of breadth confirmation would warn of worse things ahead, especially in the absence of signs that the inflation and interest rate uptrends had started to abate.
As mentioned before, as we go into 2006 we are defensive in favoring larger-caps over small-caps and Value over Growth. Like large-caps, Value has tended to outperform Growth after market peaks. And despite six straight years of Value out-performance, our indicators do not show Value to yet be expensive relative to Growth.
While the Value outlook is supported by the prospect for weakness in the first half of the year, a market recovery in the second half of the year could improve the chances for Growth relative strength. After a market-bottom, small-cap Growth in particular would have an opportunity to reassert itself, given its historical tendency to outperform in the early stages of new advances. But, of course, we must also be concerned about the market beginning to focus on a worsened earnings outlook. Earnings growth is far from negative right now, but the implications of that occurring would at first be a negative for Growth, as it would be negative for the market.
[back to top]
Industry Leadership Outlook
Relative to the market, we expect Health Care an Consume Staples to out-perform. We expect Cyclicals and Consumer Discretionary to under-perform. To derive this outlook we combined the rankings of our historical industry studies from macro-economic factors which we believe are likely to play an influential role in determining industry leadership in the year ahead. Those factors are: 1) second-year Presidential cycle; 2) declining new home sales; 3) bear market; 4) end of the Fed tightening cycle; 5) U.S. dollar weakness; 6) seasonality; 7) relative forward P/E; and 8) three-year mean reversion potential.
However, not all of these factors have an influence at the same time. For example, because we expect U.S. dollar weakness to occur in the second half of the year(i.e. as the Fed ends its tightening campaign), we did not add this factor until that time. Further, as market conditions warrant, we will update this outlook as new factors come into focus in the year ahead. In all, this composite of historical studies reflects our “best guess” of how we think leadership will unfold based on the weight of the historical evidence.
The biggest risks I see to this approach are: 1) the current cycle’s idiosyncratic factors may not be fully accounted for; 2) the timing of the factors we have included in the model may also differ from the actual timing of events; 3) the factors in this model are currently equal-weighted, but at times, some factors may dominate much more than other – particularly in the short-run and when unforeseen events are encountered. All (or any) of these could cause actual leadership in 2006 to deviate from its historical tendencies.
That being said, I do not intend for this model to be a replacement or what has worked for us in the past (momentum-based rankings combined with other indicators which update regularly with the inclusion of current data). Rather, this model is intended as a supplement for monitoring how well industry leadership is adhering to its historical tendency. In the end, we still recommend placing a heavy reliance (typical triple-weighting) on the trend – but we do not allow that one factor to dominate our models.
[back to top]
Aforementioned Macroeconomic Influences
- Second-Year Presidential Cycle: This factor remains in our model through the entire year. We have found that the Presidential cycle correlates with the stimulus provided by the fiscal and monetary authorities. Non-cyclical defensive market leadership is dominant. In fact, Health Care has always out-performed in the second-year of the Presidential cycle, followed by Consumer Staples. In addition, in a worst case scenario of second-year Presidential cycles with bear markets and recessions (there have been three cases since 1974), we found that these two sectors had always out-performed, while Industrials and Financial had always underperformed.
- Declining New Home Sales: This factor also remains in our model throughout the entire year.
As the bulk of employment gains over the last two years have been in real estate, construction, and related industries, this factor may indeed have far-reaching consequences. Again, traditional market-defensive non-cyclical leadership has prevailed, while Cyclicals have exhibited under-performance during these periods. Declining new home sales have had their most assured influence on Consumer Discretionary, Industrials, Materials, and to some extent Financials.
- Bear Market: We are keeping the bear market factor active in our model for the first three quarters of 2006 based on the four-year cyclical composite. While current tape and sentiment indicators suggest a mildly defensive posture, rising interest rates, high debt levels, and valuation suggest that the secular bear market may reassert itself in 2006. This could be the biggest facto risk to the model, although I believe that our momentum/trend models will keep us in check. For this scenario Energy, Consumer Staples, and Utilities have been the leaders – Energy has outperformed in 90% of the cases. Its only loss was in 1981-1982 when oil prices declined about sixteen percent (in today’s environment, that would be the same as a $65 barrel of oil falling to $55 – not entirely implausible). Similarly, Consumer Discretionary has outperformed in only 10% of the cases. The only case of out-performance was again in the same 1981-1982 period.
- Fed Cycle – End of Tightening / First Fed Rate Cut: Another shortcoming to our model (especially when thinking only about linear returns for the calendar year) is that useful interim detail may also be lacking (again, factors may change during the year and thus have different influences at different times). For example, one of the strongest factors influencing leadership in 2006 may very well be the Fed tightening cycle. Historically, Financials have outperformed in only seventeen percent of the cases (one in six cases since 1974) in the six months prior to the end of the Fed tightening cycle, but have outperformed in all cases in the three months prior to the first Fed rate cut.
Because we currently expect the end of the Fed tightening cycle to occur near the end of the second quarter, Financials may remain under pressure during much of the first half of 2006. But as we near the end of the Fed tightening cycle and again when it appears the first Fed rate cut is in view, our expectations for Financials will likely improve significantly. How will we know if we have seen the last Fed rate hike this cycle or just a pause? Such things can only be know for certain in hindsight, but we will use the Fed Funds future market for guidance.
- U.S. Dollar Weakness: Facing massive deficits and the prospects for stronger growth in Asia as well as other emerging markets, we think that the probabilities for another round of U.S. dollar weakness will increase as the Fed winds down its tightening campaign. Thus dollar weakness leadership is expected in the second half of the year. At the sector level, Energy, Health Care, and Utilities have historically tended to benefit from U.S. dollar weakness, while Consumer Discretionary (due to higher import prices for retailers) has tended to be hurt the most.
- Seasonality: Macro factors often underlie seasonal leadership (e.g. weather/heating oil; seasonal buying trends, capital expenditure cylces), and tend to affect the fourth quarter the most. While seasonality can easily be overcome by other factors, the pattern for Growth leadership in the fourth quarter is often a reliable one. This suggests potential tailwinds into this quarter (Q1), although we would likely view a rally as suspicious given the risks from other factors.
- Relative Forward P/E: While macro factors may set the tone for industry leadership, industry-specific factors are also key. Currently the most under-valued sectors are Energy and Health Care; the most over-valued are Utilities and Financials. This factor will be updated throughout the year.
- Three-Year Mean Reversion Potential: As a quick explanation, the market – as well as sectors and asset classes which comprise a market – tend to “revert to the mean”. (In this instance, are rankings are based on the number of standard deviations using three-year excess returns relative to the S&P 500.) Health Care is currently showing the most favorable long-term upside mean reversion potential, followed by Telecommunication Services. Energy and Utilities are showing the most downside mean reversion risk. This factor will be updated throughout the year.
[back to top]
Stock Market & Style Allocation – An Important Detail
“More important than the market’s cyclical status will be the status of market breadth, which has remained subdued and divergent over the latest rally.”
I perceive the above sentence to be very important in determining in which direction the market will break. S&P 500 valuation perspectives are more consistent with a secular bear than a secular bull. And that secular bear case will receive further support if market breadth starts looking more like it did in 1977 (a secular bear year) than 1995 (a secular bull year).
The course of the market breadth – positive or negative – would tell us a lot about whether the market will continue to meander, take a bullish path similar to 1995, or follow a course more similar to that of 1977. Today breadth measurements are still inconclusive. We have watched the breadth indicators move a little higher, move a little lower, refuse to break out, and then repeat; leaving the stock market pretty much where it has been since mid-November. It thus remains to be seen whether the breadth picture will support the prospects for another bull market leg or whether it will argue that this long cyclical advance is finally giving out.
[back to top]
Outlook for 2005 (a preamble) – and obviously a review
I will leave you this month with something I wrote in a January 2005 letter. In particular, I want to draw your attention to the comments on Type I and Type II mistakes:
“Before getting into too much detail, let me explain how I come to a prediction and what scares me about predictions. All I do to come up with a prediction is assess all the data I trust, create an interpretation, and determine the statistical probabilities of an outcome. I then go with the prediction that has the greatest statistical probability of occurring.
That is just good decision making. It is sound and it is credible and it works for me. What scares me about predictions is that when decisions are continuously made based on statistical measurements, then statistically speaking, one of those measurements has eventually got to be wrong. So perhaps I am due to get it wrong, and that scares me. Still, I would have more faith in my and my track record than a management team that has gotten it wrong for years and “is due” to get it right. At this point in my career I think it is clear that my research works and that I do not rely on luck (I hope to instill confidence in your decision to use Berkshire Money Management; I hope that it does not come off as arrogance).
Nonetheless, you can never rule out the affect of luck – both bad and good. As such, I take some precaution when it comes to investing. Generally speaking, there are two major types of mistakes we so-called “professional investors” can make. Don’t get me wrong, we can make a lot of mistakes. But the general strategies we employ can be classified in two ways.
The first strategy mistake (Type I) we can make is to be “over-invested” when the stock market turns down. The bad news there is obvious – the client loses more money than they should.
The second strategy mistake (Type II, a.k.a. the “you don’t go broke taking a profit” strategy) we can make is to be “under-invested” when the stock market moves up. The bad news there is one of an opportunity loss – the client makes less money than they should.
Either “mistake” is occurring to any person’s portfolio at any given time (excluding the occasional lucky instance of perfect timing). So, as a professional investor I need to weigh the risks and determine which mistake I, in the name of the client, am more willing to take.
Generally speaking, I lean toward the second strategy mistake, the Type II. And please allow me to explain why. Firstly, it is difficult to call the top of the market (long story short, the tools used to call a market top with precision are pretty useless. If it is done within a couple calendar quarters that is a high-quality market call). Secondly, it is relatively easy to call a significant market bottom (within a few weeks). So generally speaking, I utilize risk management (a process) as I feel the market is topping and remain patient until I am able to buy equities (an event). Said another way, an expensive market can continue to get more expensive – we just don’t know when a market will turn down. However, if we get conservative too early all we have to do is be patient to take advantage of an eventual correction.
But there is more to it than that. The above explains my investing strategy which works well for patient investors (I realize that not everyone is a patient investor). However, from the client’s perspective – the person whose livelihood this money is tied to - I don’t forget that these assets are tied to living, breathing, feeling human beings; I find that most people would rather make less (being under-invested when the stock market goes up) than lose more (being over-invested when the stock market goes down). And given what I have been able to do for clients at Berkshire Money Management since we started in 2001, I feel that I have some “mistake capital” built up where I can get a little more conservative to protect your portfolio, even if it means possibly missing out a few extra percentage points of return in the short-term. I would love to hear your thoughts on this (that’s your cue to call, e-mail, or write me).”
Last year, 2005, allowed me to build up a little more of that “mistake capital” by soundly outpacing the market. But I would still be interested in hearing your comments regarding whether you consider yourself to be a Type I or Type II investor.
Thank you for all of your business.
Respectfully,
Allen Harris
President
