Interim Investment Letter
August 23, 2015
The Friday prior to publishing the 17 Mile Annual Letter the S&P 500 and VIX closed at 2,091.54 and 12.83. In five trading days the S&P 500 declined 5.73% to 1,970.89, while the VIX more than doubled to 28.03. While far from atypical for a run-of-the-mill correction, it has been the deterioration in various third-party market models that makes the situation interesting. As I said in the annual letter my Mosaic model is almost 100% bullish, and going into the week I was positioned as such. But I also pay attention to lots of third-party market models and commentary in order to test my thinking and potentially gain an edge on something the Mosaic is missing. Due to the these third-party factors, as I will detail below, on Wednesday I decided to take down exposure by selling VRX and ZTS. What I did not anticipate was how quickly the market would decline – by Friday 50% of the positions were bought back.
While performance is negative MTD (-7.5% versus -6.1% for the DWCFT), it has been a fun month. The Media carnage on August 6th created an attractive opportunity to add to CHTR below $178; VRX’s status as a ‘hedge fund hotel’ led to a rapid decline from almost $250 at the beginning of the week to $222.19 at last close; and HTZ was puked up mid-day on Friday, down almost 7% to ~$17. As painful as volatility can be, the opportunities it creates are fantastic. And not just money-making opportunities – I get better as an investor with every single correction. Most recently I have formed the ability to look at a ‘puking’ price chart with enthusiasm rather than internal panic (I say internal because I long ago learned to not actually panic and sell into the puke). It is all mental – you have to realize who is doing the puking and understand that it is not (likely) economic, but forced. Of course this assumes your fundamental and situational analysis is correct.
I believe there is a good chance of a 2011-style mini-bear market (<=20% decline), and I want my plan for taking advantage of the associated opportunities to be clearly documented in advance…hence this letter so soon after the last one. Below I will outline the following: current status of the Market Mosaic; current status of non-Mosaic models; August MTD portfolio action; and my current strategy.
The Market Mosaic is a weight-of-the-evidence approach to market risk management that looks at economic & monetary conditions, market ‘technicals’ and market valuation. The model is predicated on the fact that historically the largest market declines (40%+) occur inside of recessions, and that outside of recessionary conditions market declines are relatively limited in both size & duration. Comprised of the following components, the model is long-term, slow-moving, and hedged against itself, designed to get portfolio exposure in line with bulk of the major market trend in place:
- Economic Conditions – Yes/No recessionary conditions
- Monetary Conditions – 10% weight
- Market Model – 40%
- SPX Price/200dma – 30%
- SPX 200dma Direction – 20%
The weighted components are assigned a score ranging between 100 and -100 in increments of ’50’, and portfolio exposure is managed according to the following two-part formula:
- ‘No’ Recessionary Conditions
- Gross: 50% min, 200% max
- Net: 0% min, 200% max
- ‘Yes’ Recessionary Conditions
- Gross: 0% min, 100% max
- Net: -100% min, 100% max
Economic Conditions. While recessions are notoriously difficult to predict in advance, they are relatively easy to identify once they have begun using an incredibly simple yet effective indicator: YOY change in the 4-week moving average of U.S. Jobless Claims. As Cullen Roche of Orcam Financial Group has highlighted ad nauseum on his blog Pragmatic Capitalism, no U.S. recession has occurred without Jobless Claims spiking by at least 20% YOY. Earlier this month I did a quick study of market returns around this indicator; and while it did not get you out at the exact top (-14% median decline into the 20% trigger), it did anticipate a median decline of approximately 20% over the ensuing 300 trading days.
At present this indicator is declining by more than 10% YOY, indicating recessionary economic conditions are currently not in place.
Monetary Conditions. With record low developed market policy rates, ECB and BOJ quantitative easing programs in place, record safe corporate balance sheets, and the safest U.S. banking system in decades, it does not take a rocket scientist to determine that monetary conditions are extremely friendly at present. The following graph of BB OAS is representative of the extremely friendly monetary conditions currently in place:
Out of deference to the ‘bear’ case, the BB OAS level is trending in the wrong direction, up over 50% YOY; but given the extremely low base from which spreads have come, I am not overly concerned at present. I assign monetary conditions a score of +100.
Market Model. The Market Model remains on a HOLD signal following a BUY, which warrants a +50 score. As noted in the annual letter, however, if the market rallies hard over the coming weeks and months it is possible the model flips to a SELL signal, as the flattish market returns over the last 6 to 9 months has dampened the momentum aspect of the model. Something to watch.
Price/200dma. As of Friday’s close the market sits dead on the bottom end of my 95% to 110% bullish range. While technically still bullish, out of deference to the market I assign this a +50 score. And to be fair, I do take into account the fact that the 200dma Direction, as highlighted below, is still fully positive. If we were in a highly negative market environment where the 200dma Direction was -100, I would rate a 95% P/200dma as -50.
200dma Direction. The 200dma continues to trend upward, and the slope – as measured by the 200dma 200-day growth rate – is over 8% at present. Score = +100.
Composite Score & Max Exposure: 65 & 165%
Sentiment. I follow five sentiment models, three of which currently sit at extremely pessimistic readings. And the only reason all five are not officially rock bottom is due to timing. In reality, all five are at rock bottom. This is extremely bullish in the short- to medium-term, as even in 2011 the market rallied significantly prior to the mini-bear market in order to clear out excess pessimism. As I will detail in my current strategy below, I believe there is a high probability we rally back close to, if not above the all-time high in order to clear out this extreme pessimism rating. From there, it can be assessed whether exposure should be reduced and/or hedges put into place in order to protect against a 2011-style decline.
Ironically, it was sentiment that drove me to reduce exposure on Wednesday. Sentiment has been negative for awhile, but as Urban Carmel pointed out last weekend, the fact that the market refused to rally hard in the face of extreme pessimism typically means the market needs to fall in order to bring sentiment down even further. This fact really bothered me, as I follow sentiment closely – and I noted this on the Trading page, rationalizing my exposure reduction. While I did not anticipate the speed with which the market would clear out excess positive sentiment, it now has with a 5%+ decline Thursday and Friday. I like that. I like that a lot.
Momentum. I follow three key momentum and supply/demand models. One has held up very well, and even after last week’s decline continues to hang in there. The other two models have been negative for awhile, and bolstered my belief that the market was at risk with sentiment stubbornly refusing to fall to a rock-bottom sentiment configuration.
Momentum models are tough because, while they keep you in the market for the bulk of major market moves, they are terrible at calling market turning points. And in long-standing upward trends, when momentum models turn negative, it is typically NOT the time to bail on the market. In other words, they must be interpreted within the context of the economic & monetary environment and the long-term technical configuration of the market. As such, typically when momentum models become highly negative within the context of a favorable economic/monetary/LT technical backdrop, it is a good buying opportunity. It remains to be seen if the recent oversold conditions of the momentum models indicate a buying opportunity – their response to a sentiment-clearing rally should be monitored closely – and if not, then we very likely are headed for a 2011-style decline.
Composite. I follow two good composite models that are very similar to my Market Model. One is highly momentum-oriented, so its effectiveness closely follows the momentum model conclusions highlighted above. At present this model is at its lowest reading in years; but in the 2005 time frame it hit similar oversold levels within a long-term uptrend.
The model most similar to my Market Model is one of the best models I have ever seen, and it is currently benign. However, similar to my model as well, in the event of a sentiment-clearing rally, it could turn negative.
Market Model Volatility. A complex volatility measure embedded within the Market Model, which has done a bizarrely good job of forecasting poor market returns even when the model itself is on a BUY signal, is encouragingly very benign at present.
AUGUST PORTFOLIO ACTION
CRC. Shortly after initiating the ‘distressed’ HTZ/CRC basket I sold out of CRC to double up the HTZ position, as further due diligence revealed HTZ is far from a stressed situation.
CHTR. As noted, on August 6th CHTR declined to below $178 alongside the Media carnage, and I sold some VRX over $250 to add to the position.
FNMAS. Due to strong returns since doubling up on FNMAS in June, and the portfolio decline this week, FNMAS grew to an uncomfortably large position, so I trimmed some this past Friday in order to add to HTZ…
HTZ. My conviction in this name has grown considerably, and I have been uncomfortable with the relatively small position size. As such, I took advantage of the 6%+ plunge to ~$17 on Friday to add materially using the proceeds from the FNMAS reduction.
VRX/ZTS. Both of these names have been ‘behaving’ poorly, so I thought they were good candidates – VRX in particular due to its strong YTD returns – for the exposure reduction move on Wednesday. I still love the names long-term, and with the material declines in Thursday and Friday, I bought back approximately 50% of the original position size.
As I have documented in bits and pieces on Twitter, I believe we will see upwards of a 3% plunge by noon on Monday, which I believe will represent a NT bottom in the market before it embarks on a sentiment-clearing rally back up to ATHs. If the market follows this path, I plan to add more VRX and ZTS; and depending on the magnitude of declines, I will add the remaining capacity to either DISH or BAC. I am strongly leaning toward DISH as it has been pounded in this decline (non-cash-flow generating spectrum does not provide much support in this market :)), but I may want to add to BAC as a trading position to sell out on the move to ATHs. I am still mulling…
In the event too many market participants are anticipating this Monday plunge and the market moves up instead, I will likely sell out of the additional HTZ added Friday as well as the re-established VRX and ZTS positions, as I believe the market makes lower lows in the NT.
More to come…