Total Return Performance of Top 100 ETFs Taken As Its Own Portfolio

Aug 06, 2013 in ETF Multi-Asset 100

Following a -2.9% loss in June (and -7.7% drawdown -- peak to trough daily),  the ETF Top 100 rallied back +4.0% in July.  Year-to-date,  ETF investors in these 100 are up +$59.8 billion.   Investors have made nearly $32 billion in SPY & IVV (S&P 500) and lost -$22.8 billion in various gold-related products (GLD, IAU, GDX, GDXJ).

The multi-asset index is a composite based on actual ETF asset levels.   It is a good proxy for what is held by ETF investors in aggregate -- and is a basic proxy for one logical look based on the proportions ownded in each.  

Exchange-traded securities are for the most part baskets of securities (some are single commodities like Gold or Copper).  These baskets represent factors and your decision as an allocator of your assets is which factors you want to emphasize in your portfolio.   The ETF multi-asset 100 was created as an objective way to try to understand what it is that investors are preferring with their investments in general.  It is not a perfect measure, but it is a logical proxy for such preferences as it weights each asset by its dollar value.

Below is the Year-To-Date performance through July 31:







91 ETFs made money in July with 9 showing losses.   Hedge funds lost -1.3% in June when most ETFs dropped and gained +1.0% in July when most ETFs rallied.   The global hedge fund indexes are a mix of many different types of strategies and utilize many different assets. 




 

Follow us on Follow etfreplay on Twitter

Tags:

S&P 500 Total Return For 2013

Jul 19, 2013 in S&P 500 | Total Return

People seem to get confused by this so let's just update the Total Return for 2013 here and explain a few things.  Below is a chart of our Free Total Return vs Price Return webpage with the same security below from a leading institutional platform.   The numbers are the same.  One of the key benefits of having an exchange is the data comes from the exchanges, not some proprietary source (which is unlike the bond 'market' -- where there is no exchange).

 

 

 

A few comments:

First,  an index does not have shareholders.   Indexes are uninvestable --- what you are investing in (be it an index mutual fund or an ETF) is a financial PRODUCT that tracks (or attempts to track) an index.

Since an index does not own any shares of the companies in the index,  companies in the index do not owe the index any cash like a regular shareholder is entitled (dividends).   This creates a difference in those securities that are investable (ETFs) and those securities that are not (indexes that aren't total return).

Now, you can buy options and futures based on an INDEX --- but those are derivatives and thereby do not need underlying securities.   It should be pointed out that ETFs are not derivatives any more than a mutual fund is a derivative.   In fact, most ETFs are registered under the very same 1940 Securities & Exchange Commision (SEC) act as mutual funds.   The basic structure is the same --- though the mechanics are different.

Some people mistakenly think that since when you buy an ETF and you are therefore getting ETF shares, that represents a derivative.  It most certainly does not.   When you buy a piece of property, you get a deed (a legal contract).   When you buy a stock, you get a stock certificate.   When you buy a mutual fund you get mutual fund shares.    These are not derivatives.   The vital difference is that derivatives do not hold the underlying securities ----  ETFs DO hold the underlying securities, which makes them regular pooled investment funds.

Now, since ETFs hold various numbers of shares, they are paid cash when companies make those distribution payments (dividends) to shareholders. It is important that you account for these distributions as when you go to rank securities,  the entire ranking process will be compromised unless you are using the same process across all securities in the list.    In fact, one error in a list of securities can impact the entire rank order and actually affect the chosen securities to be included in a backtest,  meaning you cannot have ANY errors.   

Since ETFreplay covers nearly 1000 securities, we took great care in building our processes -- and importantly, how you cross-check data.   Good thing for us is that modern databases (such as the most recently released SQL database our service provider runs on) were architected precisely to address the inherent problems in data management.    (databases have very rigid rules relative to something like a spreadsheet, which was created for the masses).  

As a sidenote, because it is frankly amazing ---- the use of Excel has been the source of two very high-profile finance-related errors in the past year ---  the London Whale incident at JP Morgan --- as well as the influential Reinhart-Rogoff economic paper used to justify the adoption of Austerity.    We maybe are less surprised that academics would makes  such an error --- but the JP Morgan case is amazing because of the mere fact that they would ever run such massive amounts of money without a more architecturally secure product than Excel.    Did nobody in the JP Morgan london office ever study the benefits of using a relational database management system (RDBMS)?    Running a database enables many levels of cross-checks and reporting that is impossible in a spreadsheet.   Caveat emptor. 

See Also:  Other Total Return Blog Posts

Follow us on Follow etfreplay on Twitter

Tags:

A Real-Money Aggregate Allocation: Update on the Multi-Asset ETF 100

Jul 03, 2013 in ETF Multi-Asset 100

The Multi-Asset ETF 100 uses the real money asset levels invested in ETFs to serve as a proxy for how the overall investment world is allocated.  By definition, this is what ETF investors as a whole actually own and in the proportions they own them.   By observing the contributors and detractors to this portfolio (with a starting value over $1 trillion), we can get a good sense of where the world at large is getting (losing) performance.

 

95 of the 100 ETFs in the index declined in June as interest rates rose and equities fell.   That said, US stocks (the largest component of the ETF 100) were down only modestly.   If you think it was a really bad month -- you probably haven't been involved in the markets very long -- it wasn't.   (And you will likely then be shocked speechless then when a real correction does eventually hit).

 

Some media attention focused on the losses experienced by 'risk-parity' funds -- it should be noted that many risk-parity strategies call for using a fair bit of leverage.  Typical unlevered risk-averse allocations were only slightly down for the month. 

 

Another interesting development is that of the TIPS market (Treasury Inflation Protected Securities).   It is estimated that PIMCO as a firm owns 10% of the entire TIPS universe.  The problem with TIPS is that they have interest rate risk (duration) --- and when you don't have inflation, you are just left with exposure to a general move up in rates.  Underweighting duration (or zero duration with cash-like securities) has been the play in the bond market.

 

Emerging markets dropped more than -5% for the second consecutive month with many popular individual country funds obviously doing much worse than the EM index.    On a higher level, do you recognize a theme in all of these examples?  The weak coming into May and June all got a lot weaker.   TIPS, Gold/Silver/Mining, Emerging Markets...   these all were making new multi-month lows in March, April and/or May prior to dropping more seriously in June.   One of the primary benefits in proactive relative strength analysis is in missing big problem areas.  Mining equities and various commodity ETFs have been an unmitigated disaster.  In some ways, it is shocking to see the YTD figures on things like GDX/GDXJ and SLV.  On the other hand, these have consistently been some of the most volatile ETFs in the marketplace.    And naturally, high volatility tends to get larger drawdowns.

 

With regard to US stocks, we have seen this all occur before.  In past years, the US market has at times been viewed as a safe haven, immune from world events.   This divergence can remain in place for sometimes longer than you think --- as investors left underinvested in US stocks see the pain in their performance comparison vs the S&P 500 and capitulate by buying more US stocks.  But ultimately,  the US market goes back to doing its thing --- getting volatile and drawing down.... and those who had believed in decoupling get burned.   Said another way --- stick to your process.

 

We compare the ETF 100 to the hedge fund index as another comparison of a large pool of assets.   After a long-run of poor performance by the hedge fund index, a small losing month shows up as a solid result and the HFRX Global Hedge Fund Index has moved ahead of the ETF 100 for the YTD period thru June.    As noted in past months,  we expect this index to outperform when global equities (and US equities in particular) fall -- as happened in June.   The very low volatility of this index helps cushion it from large drops.   The trick of course is to produce attractive returns, not just low volatility.   Low volatility/low returns can be created by anyone for (almost) free through the use of short-duration bond funds.

 

Follow us on Follow etfreplay on Twitter

Tags:

Multi-Asset Real Money Benchmark Backtest

Jun 04, 2013 in ETF Multi-Asset 100

The ETF 100 is a measure of real money performance.   Rather than pick a single index --- like the S&P 500 ---  we use a mix of assets that is weighted in proportion to how investors actually hold their assets.    We do this because this is reality --- how is the overall multi-asset-allocated world doing on a high-level basis?    This portfolio owns a lot of S&P 500, it is the biggest holding by far --- but it is not the majority.

We make as few assumptions as possible and we sought to benchmark a meaningful chunk of assets.   We think over $1 trillion is a meaningful starting point to represent a multi-asset portfolio benchmark.

Below is the path for 2013 beginning with the actual starting asset levels for the top 100 ETFs.

 

 

This benchmark allocation has made $48 billion for investors this year (total return includes dividends). It was down a bit in May and this highlights the divergence that has occurred between many different kinds of assets vs the S&P 500.    Indeed, while this portfolio has made +~24 billion on SPY/IVV,  it has lost -$14.5 billion in GLD/IAU.     That said, Gold had gone up 12 years in a row and was frankly long overdue --- investors who don't know the longer history of gold have paid the price in 2013  (see Gold Blog From December.)

As a sanity check, we also check in on the multi-asset HFRX hedge fund index.    These have drawn even in 2013.   Hedge funds have seen very poor performance for multiple years now but in relation to this broad portfolio,  the underperformance is put into better context.   Multi-asset portfolios don't go up and down with the S&P 500.

 

 

 

Lastly, think about the past 10+ years from a higher level.   Money managers who exposed their clients to international markets greatly outperformed US equities.    US-only managers went through a tough time but now have major wind at their back.    The decision of which proportions to own international vs US equities is a very important decision.    If you have been exposed to US this year --- as relative strength has guided  --- you are doing a lot better than the broad benchmarks above.   

The chart below uses our Free Portfolio Combination Tool to show the changing out/underperformance of US stocks vs rest of world.

Follow us on Follow etfreplay on Twitter

Tags:

When Rates Go Up, Which Sectors Do Relatively Better

May 31, 2013 in Bonds | Sectors

1202 days in past 10 years that TLT went down.

 

 

Now applying the 10-Year Data to Current Situation --  May 2013 (1 day prior to end of month):

 

 

 

Follow us on Follow etfreplay on Twitter

Tags:

Multi-Asset ETF Portfolio 100

May 02, 2013 in ETF Multi-Asset 100

Updating the Multi-Asset ETF 100 Portfolio performance which serves as an index of real money ETF investor holdings. The portfolio represents well over $1 trillion in real-money assets. This is a benchmark that makes as few assumptions as possible and can serve as something of a proxy for a 'market portfolio' --- we do not compare this to a single asset-class benchmark like the S&P 500. The S&P 500 is just US large cap stocks -- US stocks make up a material percentage of this index --- but the ETF portfolio is multi-asset and the S&P 500 is not.

This month saw a sharp selloff and recovery. Gold, Silver and Gold Mining ETFs have been hit hard this year. Excluding broader Materials or commodity ETFs, the Multi-Asset 100 began with roughly 9.9% weight (the sum of GLD, IAU, GDX, GDXJ as % of top 100). This is a large weighting and perhaps was warning in and of itself. The group has shown extremely poor relative strength over the past 18+ months and was quite easy to avoid.

ETF investors have made +$59.2 billion in the first 4 months of the year vs start value, much of that coming from US equities.

 

The index became dramatically more volatile in April. Nevertheless, just as in March the index closed at its high for the year on the last day of the month.

 

As can be seen in the second chart, the primary HEDGE FUND index has lagged, despite the Gold rout hitting the ETF portfolio hard. Our view is that the hedge fund index is most likely to outperform only when US equities go down.  The 27 basis point (0.27%) expense ratio of the ETF index has a huge fee advantage over hedge fund compensation.  

As a sidenote, on the worst day of the gold drop (April 15th), the Hedge Fund index had its worst day in 18 months, showing that there was some decent gold exposure even in the massively diversified global HF index.

Follow us on Follow etfreplay on Twitter

Tags:

ETF Ratio Backtest And Ratio MA Backtest For Portfolios

Apr 12, 2013 in Backtest

Ratio Moving Average backtest was a module we added in 2012.    We have since added Ratio MA Backtest for portfolios.    

Ratio Moving Strategies are good ways to do added research and help you gain comfort in how your portfolio is positioned.  We find that this type of analysis can add some perspective and can help think about some other options to fine-tune your actual entries and exits.   We outlined one way to use it here: Allocation ETF Overweight Example. 

We also recently added a slideshow to highlight our thinking on the layout of ETFreplay.com.    That is located near bottom right of the Backtest Page.  

 

Follow us on Follow etfreplay on Twitter

Tags:

Create Your Own Benchmark

Mar 31, 2013 in ETF Multi-Asset 100

One easy and free way to use ETFreplay.com is to build your own benchmark using the free combine page.

On a related note, we created the ETFreplay Multi-Asset 100 as an example of a 'market portfolio' proxy benchmark.   The ETFreplay Multi-Asset 100 attempts to make as few assumptions as possible.   We simply weight ETFs using the starting asset amounts for the year (using published ETF provider figures) and then track their total distribution-adjusted returns.  No adjustments are made for inflows and outflows during the quarter -- instead we will re-weight this benchmark periodically.   (This assumption makes almost no difference as the starting portfolio is so large (well over $1 trillion) that even many billions in net inflows don't make a material difference over shorter periods of time).

One thing to note is that equity ETFs have existed a lot longer than fixed-income ETFs,  so the fixed-income ETF movement has been playing catch-up in recent years as more and more investors turn to ETFs rather than alternate products for their fixed-income exposure.  

Large cap US stocks are the largest component of the index.   That said, large cap US stocks represented just ~37% of overall assets.   Note that we included all of the US sector fund assets in the large cap category as the sector funds (like XLF and XLK) are dominated by large cap US companies.    

An important point here is to remember that the world does not begin and end with the S&P 500.   The S&P 500 is an index made up only of large-cap US stocks.   Sometimes, this segment is the leader (something that good relative strength analysis will identify) --- but many other times, it will lag.    The point of creating a Multi-Asset index is to help put each component of the overall universe into some perspective and get away from using benchmarks that just aren't appropriate for large numbers of investors.   US large caps are very important part of the index --- but still well under 50% of the total.   Fixed-income, int'l stock and various alternatives (gold and others) matter over the long-run too. Below is the path of the overall index during Q1.

 

 

ETF investors in the Multi-Asset 100 made +$48.4 billion in Q1 versus a starting portfolio value of $1.07 trillion.   If you think the world went up double-digit percentage in Q1 just like the S&P 500, you would be mistaken.    Likewise in Q2 2013  ---  there will of course be winning and losing segments of the market.    Over the long-run, you mostly care only that your portfolio balance increases and therefore enjoys the benefit of compounded long-term returns.    And that when large cap US stocks experience a sustained downtrend, you are still managing your portfolio so that large losses are avoided.    

Q1 2013 was really the ideal environment --- at least if you are into things like risk-adjusted returns. Volatility was very low and equity returns were strong.   No matter what your benchmark is, it is obvious that repeated new all-time highs of your overall portfolio is a good thing.   The ETF Multi-Asset 100 did this and actually managed to close the quarter on its high.  

Note that a portfolio made up of 100% midcap US stocks (IJH) returned +13.5% for the quarter while a portfolio of 100% gold miners (GDX) was -18.4%.    The largest Chinese equity ETF (FXI) was -8.7% and the largest Oil & Gas MLP (AMJ) was +19.7%.     These are all index investment vehichles.   The weighted average expense ratio for the top 100 is  0.27% (27 basis pts).    The IJH (mid-cap) return of +13.5% of course includes the pro-rata expense ratio (expenses for ETFs are taken out and embedded in the market price return and would be effectively 5 basis pts for the quarter -- 25% of 0.20% = 0.0005).

We will track this index vs a hedge fund index since both are global and involve many different kinds of assets.    Below is the comparison up until March 27 --- note that hedge fund reporting is lagged by 1-2 days --- but in this case, not much happened on the last day anyway so these figures are very close to the final figures:

 

 

Think about this another way,  assume a 1.5% expense ratio and 15% performance fee on average for hedge funds in the HFRX index.    If the ETF 100 benchmark goes up +10% this year net of the 0.27% blended expense ratio,  in order for the hedge fund index to MATCH this number -- it has to first rise +13.5% just for investors to make that same 10%. In order to substantially beat that figure and actually add value for the investor, it will have to do far more than that. Given the extremely low volatility of the hedge fund industry, that seems a tall order for 2013. This leads us to postulate that the only way for hedge funds as an industry to actually add material value vs this index is if something like the ETF 100 index drops materially. The primary way for this to happen would be if large cap US stocks dropped materially. Even then, the cushion of fixed-income in the index would dilute the impact of the fall.

 

 

 

Follow us on Follow etfreplay on Twitter

Tags:

ETF Top 100 Multi-Asset Real-Money Portfolio

Mar 06, 2013 in ETF Multi-Asset 100

In portfolio management, the concept of the 'market portfolio' is discussed frequently.  As the textbooks go, if the world were 100% efficient, the only choice for investors would be in what proportion to hold the market portfolio vs the risk-free rate (T-bills).   You could take less risk by holding more cash and more risk by adding leverage to the market portfolio and so its just a risk preference.   Of course, the world is not efficient or else well-proven concepts like relative strength would not work.   Nevertheless, it's always fun to go re-visit the textbook with a real world example portfolio.  

The market portfolio theoretically consists of all assets.   There are problems with doing that in reality but in the end, we can probably create a decent proxy for all assets by making a few assumptions about what it is that dominates the movement of the overall market portfolio.

We will use actual ETF assets for this exercise.   ETFs make this so much easier to calculate because we don't have to worry about so many different mutual fund share classes that are inaccessible to different segments of people.  Trading on a public exchange makes the security available to all --- and competitive forces make it so that if you try to rip people off with high fees, your ETF will lose (or never get) assets.   If you aren't on a public exchange, then you can do all sorts of things --- charge 5% purchase/redemption fees, charge 20% performance fees etc... You can't get away with things like that when you have to compete on a public exchange (or at least not over the long-run).

The top 100 ETFs in assets (real money from real investors -- both individuals and institutions use ETFs) represented $1.07 Trillion (yes, with a 'T') as of December 31, 2012.    Using over a trillion in assets in value is an appealing figure.   We can be pretty sure that there will be some error between this trillion in ETF assets and the 'market portfolio' described in textbooks --- and that is fine.   What we are trying to do is get something that is just a proxy and acknowledge and discuss the differences.  Exchange traded products are especially appealing because assets outside of US large cap stocks are well-represented.  

This portfolio based on real-world multi asset-class weightings has a total return (including dividends) through March 6th 2013 of +3.6%,  which would be a gain of +$39 billion on top of the $1.07 trillion starting amount. That includes everything, from stocks to bonds to alternatives like MLPs, REITs, Preferreds, EM Bonds etc...

US stocks are obviously doing well and providing strong returns above the overall portfolio aggregate.  To the extent you are overweight US stocks, you are surely doing better than this portfolio.  This portfolio is getting held back by gold/silver, gold mining stocks, emerging markets -- and to some extent fixed-income.   Note that fixed-income is not losing much money, its just not making money  (junk bonds are up slightly and treasuries are down a little this year).

The return is what it is -- what might be a more interesting figure is the amount of volatility this portfolio represents and how it compares to various indexes.

Below is the same Top 100 Asset-Weighted return vs the HFRX Global Hedge Fund index.   The returns are pretty close for the year.  Hedge funds obviously own a lot of US equities -- but they are short a lot of US equities as well.   Gold is a popular vehicle in hedge funds, which would be hurting the HF index return this year --- but very likely hurting the hege fund index a lot less than GLD/IAU/GDX are hurting this ETF portfolio return.  

 

Follow us on Follow etfreplay on Twitter

Tags:

Total Return For Top 100 ETFs (Assets)

Feb 27, 2013 in ETF Multi-Asset 100

 

Visual snapshot of ETF returns for YTD period (1 day left in February).    The median for the top 100 is +3.2%.   In general; bonds, commodities and select Int'l ETFs are below the median.   US equities are above.   The largest MLP ETF (AMJ) leads the top 100 in total return (note AMJ went ex-dividend today so adjust price up by $0.51).

 

Follow us on Follow etfreplay on Twitter

Tags:

Follow ETFreplay

Archives