Matthew Crews

Long only, portfolio strategy, registered investment advisor, cfa
Matthew Crews
Long only, portfolio strategy, registered investment advisor, CFA
Contributor since: 2012
Company: Smith Patrick Financial Advisors, LLC
Thanks for the update. I'm glad it has worked out so far! Nothing fancy, just looking for value.
No problem and its freely available -- just Google:
S&P 500 Earnings Spindices
and you should find it as a xls download file.
Real taxman,
Thanks for asking but no -- I don't have insight into that. I'm not sure the cost to that information from folks like Green Street or maybe you can find something on the resource website for free. Otherwise you could pull a small comp sheet and put it together manually -- shouldn't take that much time and likely worth it.
Thanks for the reply and I don't disagree with your assessment based on what I have read. I'm also sure that I'm in the minority here but that's ok. My major concern is that I am trying to force a cube through a round hole. However -- technically there isn't anything wrong based on using Invested Capital based on actual market values. The problem is that I don't have the resources to value each & every property. And if that is the case then I don't think I am doing worse by leveraging assets on the books versus an operating income figure. It just means that MVA/ROIC picks up the property appreciation directly and indirectly via the operating income generated.
SPG is a clean example but I'm sure as I build out a group of REITs using this method I will find less enticing examples that give another angle on the quality of a potential investment.
Thanks -- well written comment!
I tried to do that with the first two bullet points. The second bullet point not having panned out and the remainder of the analysis of reasons why and what is possibly expected.
The conclusion was possibly not as concise and the "conclusion" sub heading got cut out so again maybe the note trailed off more than I had hoped.
Thanks for the comments -- agree on the concept to simplify as much as possible.
Lots of thoughts there -- thanks. I think the scare asset theme and the q factor are tied together and I have looked at the ratio which seems to tie well with long-term market valuations whether its shiller or other metric such as p/s. I use a enterprise value and ROIC on an individual company basis which also ties well with Tobin's q.
Do note that being/attempting to be pragmatic about the market doesn't presuppose I'm in or out the market. As I noted previously I was long the market despite little expected upside in 2012. This is largely to do with the fact that trends and momentum are used to help.
As for trading around Central Bank activity seems not my cup of tea. Yes -- your points are taken but they are also short-term and still trying to goose an underlying system that might/might not take to the "goosing". Debt has been shifted from the private sector to the public sector which happens to be owned by the private sector. So what if upside today has been taken from gains from tomorrow -- I don't know what to say. If it (QE etc) works and private credit pickups, wages stabilize, etc then it will be all good. Inflation data suggests we are still challenged despite 4-5 years of very easy-to-aggressive monetary policy.
Thanks again for the thoughts,
I think that it why in general those basic equations are used. Reverse engineer the expectations. If expectations don't seem realistic then one can have a greater conviction that the market is not being rational.
And the gordon growth equation is just the basic from of a discounted cash flow valuation which can be applied to an individual stock or the market at large.
As for predictions- as much as I want to know expectations -- predictions and timing will be confounded by trends. Interesting to note that over a fair amount of time earnings tend to trend upwards whereas market enthusiasm/pessimism can reallly alter valuations. So again predictions on fundamentals are helpful but psychology has its part.
Great comment -- thanks. Generally speaking that's why I like the concept of volatility clustering. Volatility will typically remain in a range and then have a burst of high volatility then settle back down. These are the periods between stability and instability.
Thanks for the comment and thoughts. Then general concept of a limiter to growth is one that I think helps people stay rational. Extrapolation is easy to do and can lead to some crazy assumptions when put into real world terms.
Dancing Diva,
I am pretty sure of the numbers. As larryl9 states -- the 40wk and 200day are similar. The 200 day did dip mid-month which doesn't get captured in the 10-month.
Rather 2013 has been significantly above trend.
First based on the timing of that post you are able to celebrate the New Year!
Its a complex system of influences. I am not sure I see the inflation yet but I think we will know more whether this recent bump in economic activity can be sustained. Clearly stagflation is a challenge if inflation picks up but productivity doesn't follow suit.
Another note is that for most of 2013 the major domestic ETFs have been significantly above trend so even a mid-teen pullback would at times put you back to the intermediate averages such as the 10MMA (my preferred average.)
As for the delta between reported and operating -- I don't have a preference outside of data availability. Are acquisition expenses one-time, are stock option expenses operating -- it just seems to get arbitrary. Bloomberg EPS might not include period pension expenses but S&P does. Again very difficult to tease that out and why reported EPS at least has some consistency that all firms must follow.
New Superhuman,
I agree -- its complex that an article like this won't nearly capture. But volatility tends to cluster between states and we are in a low vol state. A new "local attractor" likely won't be found until there is a shift in the landscape due to heightened vol etc etc (in my best attempt to recall my evolutionary biology days)
Pompano Frog,
All I can say is please review Figure 2 and Figure 5. Clearly earnings oscillate and normalization is beneficial. This isn't dogma -- just common sense. What dogma I am hearing is about central banks.
You don't agree with the data I presented -- that's fine. I appreciate the comments. And by all means you are free to relate how I've been wrong in the future as well -- if I don't beat you too it first :)
As for making it on SA for 4 years -- you might consider other rationale for that.
Thanks for the thoughtful comments. I didn't want to focus too much on reasons why momentum/trend has been strong but your right -- the market and earnings don't operate in a vacuum and the impact from politics will play its part.
One thing that I periodically look at is the private credit creation which is being supported by QE3. If private credit doesn't begin to take off -- we will be in a tough spot economically speaking. And there is little additional debt that I think the public side can take to keep the stimulus up.
Interesting times.
Pompano Frog,
I tried to stay away from CAPE on this note but only referenced it based on the dataset. I don't disagree with the comments on the CAPE and have read most sides to the argument. I think its a mistake in general to throw it out in total as averaging earnings for a cycle still holds value for investors regardless of Shiller's work in particular.
Again -- I'm trying to avoid right/wrong and just to look at the data. Since the mid-80s the capital/labor mix has shifted dramatically along with falling interest rates. One result is above average EPS growth -- above the long-term trend (and 1880s data to present does still hold for GDP/per capita). [That's why I put those charts in there.]
It is therefore no surprise that since the early 1990s that the CAPE has been above its long-term average. However current profit margins aren't set in stone nor foreign earning surplus guaranteed. Earnings can flat line like they did in the 1960s to the 1980s. If that's the case then the narrative on the CAPE will change.
Pompano Frog,
Thanks for the comments -- which there were a few. It seems if I generalize much of your first comments that the market gains will generate a positive feedback loop. Thus why its considered a leading indicator.
As for charts, they just help show the data better that 's all. Plus it helps break up the note.
As for the 19X -- its on an as reported basis which is consistent with the long-term average discussed and using the same earnings that build CAPE results. Sure on an operating basis they are lower -- but that data isn't available from Shiller for longer-term perspectives. S&P only puts it out back to 1988.
I did use operating earnings to discuss gains as I though that was more of a conservative approach to reflect gains. Sorry for the confusion as I did shift a bit there.
I linked the data source so that your can verify my work at your lesiure.
As for Central Bank beliefs I can only go with what they are doing. Clearly one can point to the current optimism as being supported by the Federal Reserve's QE3 activity.
Thanks for the comments. My thoughts might be somewhat contrarian in that QE has been associated w/ rising rates (which is argued by growth in credit) versus falling. If one thinks that credit growth includes reserves held at the Fed then a pull backin the LSAP would negatively impact rates. While contrary -- if that logic holds then tapering would be positive for treasury prices and REITS.
We shall see. Still I think there is more equity market risk than interest rate risk currently given the recent pullback.
liongterm investor,
Thanks for the comment!
Typically I like to include a company valuation in a note but focused more on VNQ this go around. I believe following the dividend yield both absolute and relative to alternatives will help keep the investment in check and when one would want to trim/add to a position.
As for interest rate spec -- I am not as sure, there is a component but its the equity version that high yield is for fixed income (which has a equity trading component). Remember that vacancies and square footage make up the "speculation"/equity component so its truly a mix. So its the economic recovery along w/ borrowing rate spreads that are driving the yield/growth
As for a good reason -- like I said I think the correction in REITS/VNQ could likely have cleared some valuation issues that hasn't occured yet in traditional equities like SPY. So the risk-to-reward has improved but still exposed to a general market pullback.
Thanks -- I'll try to keep track a bit closer of REITS and will probably do a bit more discussion on specific opportunities in the future. Also i didn't mention in the note but its amazing the flip flop of volatility where equities have been extremely quiet where fixed income has been above average.
I haven't tried to figure out a system regarding extended runs and potential tops. My rules so far have been focused on not putting more resources at those levels. It has worked out ok for fixed income but equities have been on some extended runs.
Rather I have looked to improving performance by "buying in" before end of month based on other metrics that measure breadth etc. This also works for declines mid-month -- this was extremely problematic last May (2012) for many equities and caused one of the worst months in recent history in this system at -5.X%.
Thanks for the comments!
A quick reminder -- this was a simple 6 ETF portfolio. One can get as granular as one wants. Breaking down the fixed income complex is fine. High yield has partially held up because of its short duration makeup. Also bank loans have held up. Not much else when viewing from 10-MMA including investment grade, emerging markets, tips, treasuries, muni's, mortgage back securities. Still compare that group to AGG which would been closed out at end of May?
Also -- I think there is short-term bias here. Look -- LQD has had a fantastic run! From Dec'08 - March'09 and LQD has added significant value -- only touching below the 10MMA in March'11 (and Feb'09). So now we wait until the reset finishes etc.
I kept this portfolio simple but adding GLD or DBC of recent (2-3 years) really has been of little benefit and has dampened more granular portfolios w/ a inflation-linked/hard asset component. Don't know how long it will last but GLD/DBC/TIP/VNQ all are currently under 10-MMA.
The portfolio discussed in this note is still long domestic and international equities and REITs. REITs are currently ~3% below the 10-mma so they could be sold which would push cash up higher.
I don't know if anyone puts out actual portfolio versus just buy/hold/sell indications. I have seen ETFReplay (backtesting) used to track it but it requires a membership. I personally use a custom Excel sheet with built-in rules to create model portfolios.
I believe the average is 30% cash over 10+ year time frames. Take another look at figure 4 -- details the periods where its fully invested. The average over the shown period is ~23% in cash. Its more rare that its fully in cash but that's historically because of strength in bonds. To be fully invested more often would require modified rules such as using momentum (relative strength) to purchase asset class versus hold as cash.
Thanks for commenting. Would you mind giving an example? We can look at three major buckets: equity (just mentioned), fixed income, and inflation-linked hard assets (reits being likely more correlated w/ equity).
What would you suggest with GLD or maybe just broad exposure to commodities through DBC?
Agreed on the predictability of income. And clearly it is a strategy for some but not others.
Saying that -- the system w/ an eye towards minimizing drawdowns does present some predictability from a total return aspect. However the "mental" framework has to change.
This is a conceptual challenge right? I don't disagree except that in general the markets are trending higher so therefore the mean is trending higher? And when its not -- your either out or getting whipsawed (like GLD recently or possibly EEM) There are times that trades are done at losses which is a frustrating process and means it is not for everyone.
The current period for Fixed Income is unique and the dislocation is rather large. However, fixed income is typically above trend for 80%-90% and a bit lower for longer duration fixed like TLT. So these should not trade to often.
One thing not mentioned in this note is that I track the % above/below the mean and the average % when above/below the mean. So right now Domestic Equities have been trending above the LT average so I have been less aggressive with "new" cash allocations although I have not benefited as much from the recent rally. This is one way I deal with your point on mean reversion
Lastly as mentioned in a previous reply -- I have looked at and review frequently a risk on risk off rolling window to help "speed up" the system against turns in the market (up and down). Lately the risk off window hasn't shown its face much. The last real window of risk off (for equities) was last October - November. Regardless I believe this rolling roro window does help speed up decisions and improve performance. A good example would have been last January 2012 when the markets trended above (crossed over) just after the 1st so the system missed January's rally. The roro window was risk on so based on that extra step you would have captured more upside
Thanks for the comments!
I think the crossover idea is similar but I haven't done any work on the 200 day/30 day except for what I see when looking at price charts. I do believe that you will see more crossovers and trades with that system. I believe the 9-MMA is something that I need to do more work with as an investor friend did a lot of research and felt that had some compelling results.
Something I have done with the equities side is use a window (risk on risk off) and when the system is risk off -- sell at crossover or 1/2 or something. If the window is risk on -- hold even if crossover happens mid-month. As for fixed income -- something against the 10-year yield might work ($TNX at stockcharts) but I don't have anything in place.
Thanks for the comments!
Thanks for commenting!
I periodically check the holdings of GTAA and have concluded there is a fair amount going on in the background in that ETF and not sure that it follows the Ivy strategy in its simplified form. My own real dollar experience is similar to the model and not GTAA however it also depends on the allocations and investment choices.
For instance the use of SPLV versus SPY or VTI instead of PKW and IJS. Or VEU instead of EFA (or PID) & EEM (or EELV) or AGG versus a combo of MBB, LQD, HYG or adding TIP or DBC or not DBC etc etc. Again the simpler systems even with 30% bond allocations if sold at the end of May should be doing fine and much better the performance of GTAA.
Also I would note that since last summer -- outside of Sept-Nov 2012 the trends have been reasonably strong both up & down which should help the performance of a trend system. Its the choppy-ness w/ short-term bursts of volatility that the system struggles in like we saw in 2010 and 2011 versus the 2007/208 down trend.
Thanks again
As for IC I use a quick method of TA - NIBCLS (Operating Method)
There are debates on operating cash, goodwill, operating leases etc.
I have decided that a view from the investor should include all cash and goodwill to make the ROIC more conservative. I will add R&D, Advertising, and/or Operating Leases when they are >10% of book calculated IC. That's my own methodology currently. In the past I have "capitalized" stock options but not recently.
My belief is that there are nine ways to Sunday but the key is keeping the approach as consistent as possible. My calculation might not be comparable to someone elses or a company's but it should be across the rest of my universe.
Thanks for the comment. I have found that the EV/IC metric using the 5 year average has a strong tendency to match up with a more thorough DCF calculation. It has its over/under but good enough for a screen and fair value review.
Look again at UNP. The current EV/IC reflects the improvement in margins and is significantly above its 5 year average. So if structural changes in UNP and the industry are driving those margins then the new multiple is likely closer to fair value and the 5 year average will come up to it.
Thanks for the comments -- interesting points. It is definitely on the radar, and if it does hit the $60 range based on significant pessimism at trough times then it surely would be an opportunity to buy.