Everything Is Correlated?

Oct 01, 2011

Myth "there is little room for active managers to add value because everything is correlated"

Tags:

Recent Rebalancing Subsequent Quarter Results

Sep 29, 2011

As we head to the end of the quarter, the idea of re-balancing is back making the rounds in the media and by fund companies.

When treasury bonds outperform the S&P 500, you will often get a rebalancing by large institutions. We saw this at the end of June 2011. Let’s take one type of look at how this rebalancing trade has worked in the subsequent period (quarter).

We used the 5-year treasury bond index and S&P 500 since 2000. 5-Year Bonds have outperformed stocks by at least 3 percentage points in 16 calendar quarters since 2000. In the next quarter, the S&P 500 tended to be very volatile, dropping 9 times with an average -10.6% loss. The other 7 times it rallied on average +11.0%. So in case you weren’t already thinking so --- don’t be surprised to see a big move in Q4 one way or the other.

At the end of June 2011, the story was that funds rebalanced towards stocks. We have always questioned this type of conventional wisdom and wrote instead about the opposite --- that there was relative strength in bonds. As it turns out re-balancing to stocks in June was a poor decision --- stocks suffered significant double-digit losses in Q3 2011 and treasuries continued up. Indeed, the spread between being long ~5 year Treasuries and long S&P 500 is 17.7% since June (with a few days in the quarter left).

Large funds would argue that the trade is not an attempt to be a money-making move --- just a long-term adjustment back to neutral. That is fine -- but others seem to take it and imply its a useful strategy. As we noted previously ( Rebalancing Theory & Nuances ), the value of re-balancing is not a significant source of value-add. The ‘allocate and re-balance’ conventional wisdom is in our view materially overrated by the buy-side community. The results (below) are quite wild so there isn’t a lot to be drawn bearishly either – it’s just more a matter of the fact that this does not have the characteristics of what constitutes good supporting evidence of a strong strategy (lowering risk and/or materially better return relative to that increased risk).

 

Here is sample of recent results:

 

New Product, Check It Out:

New Allocations Module Blog

Tags:

Low Interest Rates Hurt? Reality Check

Sep 24, 2011 in Bonds

One thing we hear is that today’s low rates are bad for savers that own bonds.    What?    Last time I checked, a nice return is good for a saver.

 

 

Bonds do not return just their stated yield-to-maturity (unless it’s both a zero coupon bond and held to maturity). As any credible fiduciary is well aware, there has been significant appreciation in the prices of various bonds recently.    Any bond-market investor is sitting on a strong return right now.

Complaining about low yields is a bit odd.    It is a lesser version of having owned Apple all the way up and then complaining that the market cap is so high it leaves little room for more upside. Reality check -- already made an outsize profit.   Low yields now mean that bond investors just made a lot of money on the capital appreciation side. People barking at Bernanke for keeping rates low and not allowing savers to make money is almost comedic --- the YTD total return of IEF is +14.7%.    Maybe savers should celebrate by getting their duration down and be happy with making 2 years of 7% return in just the last 6 months.  Perhaps they should consider gradually moving towards lower credit-quality bonds --  where rates have gone UP recently, not down.    Markets are dynamic, smart rotation is necessary to achieve a good return.

It seems to us that a stunning amount of investors just don’t understand the bond market. It is not a monolithic market that can be summarized with mentioning the current 10-year treasury yield.   Bonds with very long maturities are risky --- they are often more risky than a typical stock index.   Look at EDV or ZROZ or TLT--- these funds are extremely volatile. Just because they are labeled bonds absolutely does not make them conservative investments.Conv ersely, ultra short-term bonds have no sensitivity to interest rates -- they mature so fast that higher rates means higher re-investment rates for these maturing bonds --- these funds have obviously not appreciated much this year.

We like focusing mostly on the intermediate duration segments.   You get all the non-correlation benefit when things like equities are not attractive (like recently)--- you also get a decent lower-risk return while you wait for this equity market downtrend to run its course. And perhaps most importantly, short and intermediate bonds won’t get the large drawdowns when TLT and other long-duration bonds eventually tank (yields rise).

See also:

Yield To Maturity Charts Don't Tell The Full Story

Tags:

New Module: Allocations [Beta]

Sep 19, 2011

We have added a new module to ETFreplay.com that expands the functionality in a logical way. It is located on the My Account tab:

ETFreplay.com has advanced methods for screening and backtesting various relative strength and moving average strategies. The ETFreplay.com community runs thousands of backtests and it makes sense to see how members may be able to help each other with shared research ideas.

If you think about portfolio management, it doesn't matter what your process is --- in the end, when each day closes --- you have a portfolio filled with securities. Even if your portfolio is mostly cash, that is still an allocation. So our new Allocations module allows you to go to the final step in a process -- that is, take everything you know --- from fundamental to quantitative etc... and create an allocation. Once set-up, we will automate the tracking of the total return -- as well as calculate some of the most important items --- such as your portfolio volatility and portfolio drawdowns and sharpe ratio etc...

Importantly, you can do this privately if you would like --- or you can place the allocation you create on the public Allocations Board. You create your own customized fund symbol (ie, ETFRS) to identify the allocation. You can comment on each others allocations or choose to have comments turned off. You can keep watch lists of various strategies (from those which have chosen to be public). You can delete or create new allocations. We have tried to design it in a way that simply helps you to make better portfolio decisions. And as we've done with all of our applications, we will make the module significantly better over time.

ETFreplay.com has provided research applications that aid in the investent process --- but we simply can't model every possible variation of your requests into a few clicks. After you have researched and tested some ideas and have weighed all the variables and then have decided what to do, what we can do is capture the final allocation that you enter.

Note that we have installed some rules that may seem a bit rigid at first --- but these rules are necessary as we do not want to have this become a short-term trading competition. This is about portfolio management, not the latest earnings speculation or the macro news-of-the-day. The graphic below is one way to visualize the process as we see it.

 

 

Note that you can set an allocation to begin and it will use that days closing values. You can update on any day and it will then save the result of the historical fund return and subsequently combine that with the return of the new allocation. We will capture the total return of all securities (adjusting for dividends & cap gains distributions) so that your income securities are properly tracked. We know that relative to a real account, sometimes you will get a price that is different than the closing price -- but this is an allocation app and its purpose is to show various types of portfolio strategies --- not to perfectly mimic intraday trades.

Over the long-run, allocation moves are the overwhelmingly dominant factor in portfolio performance. Sometimes you may get into an ETF at a price $0.25 better than the closing price --- but this will be partially offset by the times your entry is worse. In the long-run, this is a lot less important than other factors.

We have been in beta testing on this the past few months and have worked out many of the details. We plan to add various features to this module over time. We are calling this a BETA product for now as we may have some plumbing adjustments necessary depending on how many people find use in this module and will surely have to allow for the scaling of data should many people like this feature. So please give it a try and tell us how you like it -- it is included in your existing subscription so there is no additional cost. We are excited about the potential this new application has should the ETFreplay.com community begin to share ideas and learn from each other not just through talk --- but through portfolio allocation actions.

Regards,

ETFreplay.com

Tags:

High Correlations? And what?

Sep 15, 2011 in Correlation

This is getting silly.  Portfolio managers around the world are complaining about high correlations.   The truth is that things that are somewhat correlated to start with (stocks) will see their correlations go higher when volatility rises -- it's not a mystery, its mathematical.   This has happened many times and while people are now saying ‘all-time high correlations’ – this misses the much more important point.

Crises cause securities to have larger than normal moves --- and correlation is a measure of how far one security moves in relation to the movement in another. Essentially, when one security has a big move vs its history, does the other also have a large move relative to its historical movement?   

Correlation is somewhat of a tricky topic.   Something may move very little but still be highly correlated to something that drops a lot.   Look at the utility sector (XLU) vs the S&P 500 (SPY).   XLU has a 0.89 correlation to SPY over the past 60 days.   But XLU is slightly positive so far in Q3 while SPY is currently -9.5% QTD.

Let's do a portfolio example just to see how better to think about this.  If I take 85% ultra-short term bonds (t-bills), but blend it with 15% KBE (bank industry etf) and I compare this to 100% bank index, the result is like this:

 

You can see how the cash has diluted the loss to just -3% this quarter.   (KBE is -21.1% during the same period).   The correlation has been 0.99 - 1.00.   From this standpoint, a correlation of 1.00 has told you nothing about the true risk in this case.   The correlation is so high obviously because the combination of KBE and t-bills moves a lot whenever KBE moves a lot.  It doesn’t move a lot on an absolute basis --- but it does move a lot relative to its historical range of movement.    Easy enough.

 

So back to using different securities --  utilities high correlation with the S&P 500 hasn't prevented it from going up recently (slightly).   Utilities have gone up despite the correlation between these two rising.  Thus, when portfolio managers are complaining about high correlations – they are just really complaining of the natural result of high volatility.

Earlier this year we created a simple – but what we think is more useful --- application than your standard correlation matrix.   The ‘Down Day Stats’ module measures how much various ETFs move on just the days a benchmark drops more than X%.

 

Let’s use the industry ETFs (equities only).  We can use SPY to show all down days this year.   You find that defensives like Pharmaceuticals, Food & Beverage and Telecom outperform on down days.   Energy and mining do the worst.

 

But then flip it around, use the inverse S&P 500 (SH) to show what happens on up days.   Energy and mining are best and defensives are worst.   They are all highly correlated – so you must consider volatility #1 --- and then think about correlation #2.   It would be nice if lots of equity securities were not correlated.  But the fact is they are and it gets worse whenever there is a crisis. If you want a market that is less correlated, then you need to have an extended period without any volatility storms.

 

Summary:  If the market goes down, volatility will rise and this will push correlations up.  If the market goes up, the reverse will occur.   How many times have we seen this over the past 15 years? Many.

 

Why would it stop in 2011?

 

See also 2010 Posts On Correlation

 

Tags:

Rebalancing: Theory & Nuances

Aug 23, 2011

If you think about the last few years, we’ve had stocks move to all-time highs in 2007, a big bust in global stocks in 2008 and a big move back up in 2009-2011 --  and then a sharp correction here again in Q3 2011.   

Something like rebalancing a portfolio during these moves was positive for return overall using the simple assumption of using just stocks and bonds.   However, as with nearly everything in portfolio management –there are caveats and nuances that surround every commonly accepted 'rule.'   

For example, a 60-40 stock bond portfolio picked up good return by periodic re-balancing over past 7 years.   It can be debated which is the best re-balancing rule ---  but it should be pointed out that rebalancing, while moderately useful – is not  a large source of value-add.     ‘Allocate and rebalance’ seems to be the most commonly advocated strategy among firms who are just happy to earn fees off your money.    What is often not openly disclosed is that rebalancing during a bull market (like re-balancing in the second half of 2009 -- away from stocks -- was in advance of large move up in 2010-11) hurts returns and offsets a portion of the value picked up when rebalancing works.   

Now assume your allocation was 55% Stocks, 35% Bonds and 10% Precious Metals ---- you would actually be better off having never rebalanced.    That is, the bull market in precious metals was more than the net benefit picked up by rebalancing other parts of portfolio (due partly to the cost of rebalancing in 2009 offset by the benefit of rebalancing in 2008).     Maybe this is obvious – maybe it is not.  But the point to remember is that rebalancing is something that can help somewhat over long-run --- though its total effect is likely overrated by most given what you read by financial firms that fail to acknowledge that this isn't a robust, universal strategy.    The REAL value-add is in portfolio management is having exposure to market segment leaders (this year that means stocks and commodities in Q1 2011 and precious metals & bonds since then). note: We are of course assuming you are always still staying within your broader risk tolerance when we discuss pursuing market leaders.

Note that our basic Portfolio Allocation Backtest App has a drop-menu for rebalancing rules.   To add some fun to this example, we use Berkshire Hathaway (BRKB) as the equity 60% -- and Bill Gross’s Total Return Mutual Fund (PTTRX) as the fixed-income 40%.     We compare this to a 55% Stock Index (SPY), 35% Aggregate Bonds ( AGG) and 10% Gold (GLD).    The rebalancing helps the Buffet-Gross mix --- but hurts the Stock-Bond-Gold allocation over this particular time period.

 

 

Tags:

Moving Averages Have Confirmed Trends

Aug 17, 2011 in Country Funds | moving average

Equity Indices have confirmed the loss in relative strength made a few months ago -- snapshots below:

Country Funds (note the different types of countries that were already in downtrends at or before last month-end:   Canada, Australia, China etc...)

 

 

Here are 25 Very Common Indices (of course, bonds and gold have been in bull markets):

 

Tags:

Slightly Worse Than The Flash Crash Week

Aug 06, 2011

First, a look back at this past week.   While the flash crash of 2010 was a memorable one, this past week saw many indices go down more than the week of the flash crash.

 

We have made the point over and over and over again in this blog (and indeed the underlying thesis of the entire website) of how tactical allocation is what really matters to portfolio performance.    Picking stocks in an environment like August 2011 or May 2010 or Q1 2009 or 2H 2008 is akin to arranging deck-chairs on the Titanic.  Material larger tactical adjustments are the key decisions.  ETFs make at least the implementation aspect far easier for the investor.   

Bonds have been in a bull market for 4-5 months now.   We’ve written about the relative strength bonds have been showing --- but the parabolic moves of bonds here are unsustainable and lowering durations  is a prudent move, relative strength or not.  

The other big point we want to highlight is that of focusing on your Sharpe Ratio (volatility-adjusted returns) rather than just a traditional index.   If you are going for good sharpe ratio – and you should – then you will quickly see that you are not going to be able to get a good sharpe ratio if your portfolios standard deviation is too high.  Return is of course the #1 factor --- but remember this:   the higher the vol, the higher the draw  (as in drawdown).   

When we released the backtesting applications in 2010, we fielded some questions about how to implement the concepts --- which we then enhanced the apps with solutions to address these.   At the same time, we repeatedly discussed how the real important part was the higher-level decisions.   Implementation techniques are of course important – and we strive to be as specific as we can by using well-defined rules for the applications – but in the end, this month again shows that implementing the concepts and even just getting it ‘generally right’ is plenty good enough.

After a few flattish consolidation months, it may start to seem frustrating --- but this is how markets are – lots of low returns and then some non-linear bursts of movement.    Now after the steep losses of the market recently, getting bond market (positive) returns should sit just fine.   Indeed, the updated performance of the sample portfolio is +12.3% vs -4.6% for the S&P 500 YTD.   That is using ‘first day’ picks specified a full day in advance.   Using actual picks the return spread is a bit higher.   Similarly, the other simple portfolio we used as an example (last November) – called ’01 Beta Strategies’ is currently +1,590 basis pts better than the S&P 500 (+11.3% vs -4.6%) using first day picks.  

We would like to emphasize though that your time and effort should not be to obsess over how to beat the models -- or even track them perfectly.   Models always have some assumptions in them that will keep them from being perfectly realistic.     In the end though, it should just be to get it 'generally right' as often as you can -- and good models can clearly help in that regard.  

Lastly, we would highlight that research is an ongoing process, continual enhancing of strategies is important --- small improvements add up over time.    ETFs are an empowering innovation -- there are always a few very interesting new products that come out each year.   This industry is still in the early innings of its development.

 

 

 

Tags:

June-July ETF Money Flows

Aug 01, 2011

As June ended, the story was the end of QE2.  The conventional wisdom at the time was that there would be no incremental buyer of Treasuries as the Fed ended its purchases on June 30th.    The story of the day was that stocks were cheap relative to bonds.  By the end of the month, the story was about the debt ceiling vote by Congress.   This will come and go and we will be on to the next employment numbers and european debt and whatever else comes along.  However, having a valid investment process that assesses conditions and adapts to them will serve to keep us on the right side of any intermediate or extended trends that appear.

Put yourself in the role of someone who stands in front of an investment committee – or a consultant that must travel around and tell the strategy of the firm.    It is very hard to come out with an argument that can defend a large overweight in treasuries these days.   Indeed, trying to find the reasons to buy a low single-digit yield can get you laughed out of the room.

So what happened during July?  Intermediate treasuries (IEF) rallied +3.2% and outperformed the S&P by 520 basis points (S&P 500 was -2.0% for month).

We think this is where professional investment advisors tend to really excel: in finding credit securities that are reasonable return for relatively low risk.   Intermediate corporate bonds as measured by the plain vanilla Vanguard Intermediate bond fund  (VCIT is the ETF version of VFICX) rallied +2.3% in the month, the 7th best return in a calendar month since 12/31/02 ( 103 months).  Here is the range for this index over that time period lined up from low to high:

 

 

There were of course other interesting things happening --- the Swiss Franc ETF (FXF) and all longer-term  bonds had strong months.   There were some tradeable longs in the Australian dollar (FXA) and elsewhere across the ETF landscape.

We believe that altering allocations --- even when using plain-vanilla index funds (which by the way are ultra low-cost and highly liquid) -- is a more powerful strategy than trying to understand business fundamentals of a few companies better than others.    To the extent you can find unique products that beat indexes, all the better.   But it is the basic big picture allocation that is the important part.    Do you think a $50 billion pension fund manager can nimbly move assets around on a monthly basis?  No.     Do you think that sub-second high-frequency hedge-fund trading signals affect calendar-month returns?  No.

As we’ve highlighted before, you can now trade index baskets for no commission and a penny bid-ask spread.   We live in a time that has never been so investor-friendly.

Follow ETFreplay On Twitter

Tags:

Total Returns And Footnotes

Jul 21, 2011 in Total Return

Total return is a concept that is surprisingly misunderstood.  We get emails asking why does our moving average not match Yahoo or Tradestation?   

Most Internet data sources and brokerage software platforms don't track total return -- yet total return is how all index returns are stated.  In 2010, the SPDR S&P 500 index fund (SPY) was not +12.8%, it was up 15.1%.  Vanguards investment grade bond fund (VCIT) was not +5.1% in 2010, it was +10.0% (and had some nice tactical swings throughout the year).   The difference was distributions (which come in 2 forms:  dividends and capital gains distributions).  

One common thing we see is for various people to compare their performance to the price-only +12.8% and then footnote it saying 'dividends excluded'?   To us, this is just as bad as mutual funds that claim the expense ratio is 1.2% and then footnote it saying you will be charged a 3% redemption fee if you sell the fund in first 5 years.   There are no hidden fees with ETFs -- no sales loads, no purchase or redemption fees, no 12-b-1 fees.   The 'A-class' 'B-class' 'C-Class' 'H-Class' 'I-class' mutual fund system is non-sensical in the ETF world --- there is only a single class of an ETF.     

 

Here is another one: the Vanguard 60-40 stock-bond balanced fund returned +20.1% for the 10 years ended June 30th, 2011 [1]

[1] excludes dividends

 

Well, a large part of owning a bond in the first place is the coupon.   As it turns out, that index fund (VBINX) was +59.7%  INCLUDING dividends.  So it is entirely disingenuous to compare to a number that is one-third of the actual index return. 

 

 

While this is all obvious to some -- it is clearly still not understood by many.

We created a free Comparison Tool to make it easy to view this concept as we feel the only REAL way to truly understand something fully is to interact with it through an application -- rather than just read about in a paper or on a blog. 

Try a few out.  Be aware that even if its not a large dividend payer, any capital gains distribution will also affect the return.    

Bond ETFs:

IEF

TIP

HYG

LQD

 

Others:

AMJ

RWX

XLU

PPH


 

 

Tags:

Follow ETFreplay

Archives