Today I want to consider a counter-intuitive idea: trading (nay investing) with a very long holding period (think decades, i.e. no stops). Trading horizons today seem to have shortened with social-media inputs, easy access to data, powerful technical analysis tools and low-cost, lightning-fast trade execution. Perhaps this opens the door to counter-trend thinking.
In his 2017 letter published recently, Warren Buffet shows the four worst draw-downs in Berkshire's stock (ranging from -37% to -59%) and makes the case for holding a position through such draw-downs (see Figure 1). Now BRK may have recovered from every deep draw-down, but not every stock with a deep draw-down will recover to new highs. Alternately, a stock could be locked in some seemingly irreversible long-term downtrend (think GE or IBM), so some stock rotation mechanism is desirable and indeed, the BRK portfolio has changed over time. Thus, holding through 60% draw-downs can be difficult to do. Hence, I will try to explore if understanding the context for these draw-downs can give clues that will make long-term investing easier. I will also suggest a long-term stock ranking tool to make portfolio changes easier.
Figure 1: I have taken the liberty to reproduce the Berkshire draw-down table from Warren Buffet’s 2017 letter.
The Context for Draw-downs
Berkshire's draw-downs have occurred during major recessions in the US, or more generally, during a significant slow-down in corporate earnings (see Figure 2). In Figure 2, I used the NBER recession dates (see Reference 1 below) and used the decline in corporate profits from their most recent all-time highs (see Reference 2 below). The corporate profit drop reported in Table 1 is either averaged over nearby intervals, or I used the largest drop around the time of the declines in Berkshire stock.
Figure 2: Relating BRK draw-downs to NBER recession dates, market gains or losses and drop in corporate profits.
Observe that the major declines in Berkshire stock did not occur in isolation. The two worst declines occurred around the time of major recessions in the US. All were accompanied by a decline in corporate profits measured using Federal Reserve data series (see Reference 2 below). The 1987 October crash cratered most stocks, so this one data point is not surprising. The 1998-2000 decline was the most puzzling since it occurred at a time or rising stock prices, but we can understand it better by looking at the trend in corporate profits.
One lesson from this table is that long-term investors need to monitor macro-economic conditions, since recessions are likely to see steep draw-downs in stocks.
The corollary here is that if you could avoid being in the market during major recessions, you could a) avoid major draw-downs and b) find low-risk entry points into the market (or stocks) having protected your capital. Naturally, you could use technical analysis indicators to avoid major market declines.
The essential conclusion is that you should sit through all other draw-downs, however painful they may be, if the economy is not in recession or corporate profit indexes are making new highs. Since the NBER recession identification is usually too late to be useful, a better alternative to identifying recessions may be found here. You could also monitor the $$IPI industrial production index in the StockCharts.com economic data suite.
Berkshire Declines Are Roughly Connected to Corporate Profitability
In the previous paragraph, I looked at only the four largest declines. Here I expand my analysis of annual declines in Berkshire stock and try to relate it to declines in corporate profitability using Federal Reserve data (see Reference 2 below). The two series are not on the same time scale, so I had to approximate the value of the decline in profits used to create the chart. There seems to be a definite link between corporate profitability conditions and drop in Berkshire stock (see Figure 3).
Figure 3: The largest annual declines in BRK occurred when US corporate profits had declined from their most recent all-time highs i.e. overall business conditions in the US economy were difficult.
The lesson here for investors is that they should also monitor broad-based US corporate profit indexes in addition to the recession scan I suggested earlier. For example, you could also monitor industrial production data (using $$IPI) as an alternative check on the health of the economy.
Is CAGR Growing or Slowing?
The extraordinary growth in BRK stock valuation has slowed somewhat, perhaps due to the sheer size of the company. Long-term investors should be aware that the compounded annual growth rate (CAGR) seems to be slowing from the torrid pace of the first twenty-five years. Today's large cash position and the need for ever-larger acquisitions could lead to further decline in CAGR over future decades. The lesson here is that the long-term trend in CAGR should also be monitored for constructing your portfolio. For reference, the long-term CAGR for the S&P 500 index is just a hair under 10 percent.
Stock Selection via Monthly Chande Trend Meter
Berkshire itself has changed its largest holdings over time. I have no way to know how they decided to make their changes. I propose as much simpler technical analysis approach: using the monthly Chande Trend Meter (CTM) rankings on a universe of large-cap stocks such as the S&P 500 Index.
First, I will show you the current monthly CTM ranks for the 14 largest holdings in the Berkshire Portfolio (see Figure 4). The data series for the new GM and Phillips 66 are not long enough to permit a calculation of monthly CTM data. The average monthly CTM for the stocks in the BRK portfolio is about 89 and ranges from 80 to 97. Remember, my criterion for all time frames is that a stock with CTM > 80 is trending up strongly. So, the stocks in the BRK portfolio are reporting strong trends, though this is to be expected after the recent market rally. I do not have the values of the monthly CTM at the time BRK acquired the stocks, but my analysis of their Apple entry suggests they are buying during declines. In any case, we can use the Berkshire portfolio CTM range for reference.
Figure 4: Stocks in the Berkshire Portfolio ranked by monthly Chande Trend Meter. The average value is approximately 89. The broad guide line is to analyze stocks with CTM > 80, and the BRK portfolio seems to follow this approach. CTM values need quite a bit of historical data, so missing values mean the data series may be too short to calculate the CTM.
I next ranked the stocks in the S&P 500 index using the monthly CTM (though you can construct a list of any stocks you want). The scan code could not get any simpler (see below). The results of scanning the S&P 500 index stocks using the monthly CTM is shown in Figure 5 below. The top 5 stocks are Boeing (BA), Nvidia (NVDA), MasterCard (MA), Amazon (AMZN) and DXC Technology company (DXC). If you look at their charts, you will agree that they have had good moves. You can choose to be a momentum player by choosing to investigate stocks at the top of the list. Or, you can choose to be a counter-trend (read value) player by choosing to investigate stocks at the bottom of the list. Naturally, you can choose any stocks to rank via your own stock list.
Figure 5: I ranked the stocks in the S&P 500 universe using the monthly Chande Trend Meter. Such a ranking process can help you explore ideas for modifying your portfolio over time.
One must be bold to buy and hold and be willing to tolerate very large losses in the hope that the stock will eventually make new highs. Though BRK may have recovered from every steep loss, not every stock with a steep loss will recover to new highs. It may be a bit easier to tolerate losses if they are part of a major market pullback due to recessionary conditions. However, when the pullback occurs during a broad market uptrend, even greater conviction is needed to hold on to the position. A look at corporate profit time series could help set the broader context for stock performance, making it easier to hold on to losing positions. One can also weigh the potential for recovery and market outperformance (i.e. CAGR) as part of the decision to buy and hold. You could also use the $$IPI industrial production index as an alternative measure of macro conditions. (A macro-economic conditions monitor is available at ETFmeter.com along with key market exposure measurements.)
You can use the Monthly Chande Trend Meter on your favorite stock list or market index group to quantify long-term trend strength. The scan code is shown below and could not be simpler. This scan could help you modify the portfolio over time.
Thank you for looking in and I hope you will find some food for thought in this counter-intuitive analysis in favor of very long term investing. I hope you will be inspired to subscribe to the blog using the easy link below.
- National Bureau of Economic Research dating of recessions in US http://www.nber.org/cycles/cyclesmain.html
- Corporate profit time series: U.S. Bureau of Economic Analysis, Corporate Profits After Tax (without IVA and CCAdj) [CP], retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/CP, February 27, 2018.
- Hamilton, James, Dates of U.S. recessions as inferred by GDP-based recession indicator [JHDUSRGDPBR], retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/JHDUSRGDPBR, February 27, 2018.
Scan Code for Long-term Investors
// Long-term stock ranking via Monthly Chande Trend Meter (CTM)
// Investors look for CTM > 80; Berkshire Feb18 average is ~89
[group is SP500] // select any group or list you want here
rank by [Monthly Chande Trend Meter]