CEF vs. ETF Arbitrage in the Real Estate Sector

Conclusion

Closed-end funds (CEF) and exchange-traded funds (ETF) that invest in commercial real estate equity securities exhibit a high degree of correlation (r^2=62). This high degree of correlation is despite dissimilarity in their real estate security holdings. Most surprising is the very low trading volume of CEFs versus ETFs (despite some comparability in asset size and outstanding shares) and the high degree of concentration of similar real estate securities held in the ETF portfolios.

click to enlarge

One potential trading opportunity between the ETFs and CEFs may be inherent in the tendency of CEFs’ NAV discounts to contract and expand in market swings upwards and downwards, respectively, in contrast to ETFs that are designed to trade at NAV.

Due to the similarity of ETF portfolio holdings, there may be greater arbitrage potential among the real estate ETFs than between ETFs and CEFs. Clearly, given the relative trading volumes, ETFs are a superior trading tool to CEFs for real estate.

Other Investment Considerations

Independent of any arbitrage potential, real estate CEFs currently may be more attractive than ETFs for those investors with intestinal fortitude, a modestly cheery investment outlook and a 6 month investment horizon. This would be particularly true in light of real estate CEFs’ abysmal price performance in 2008, year-end tax loss selling (further depressing prices), unprecedented yields, even after anticipated reductions, hard assets values and “discounts” on “discounts”. (See my article, “Real Estate CEFs 'Crushed' YTD 2008.")

Real Estate CEF vs. ETF Comparisons

The accompanying table illustrates the funds selected in each of the fund categories (CEF vs. ETF). The key selection criterion for each was a significant level of average daily trading volume. The unweighted average monthly price change was employed to construct the comparison graph above.

As the table below illustrates, the average ETF is somewhat larger than the average CEF. Yet, CEFs have on average a greater number of shares outstanding than ETFs. More curious is the average daily trading volume. The average daily trading volume of the ETFs is 15 times greater than that of the CEFs. Even excluding the largest ETF’s trading volume, the trading multiple would still be over 5 times.[1] The percentage average daily trading volume of CEFs is a fraction of average shares outstanding: less than 1% for CEFs versus 23% for the ETFs.

On average year-to-date, CEFs’ share prices have declined more than ETFs: 75% vs. 54%, respectively. The average percentage discount for CEFs is 29% vs. 5% for ETFs.

CEF vs. ETF Holdings

Despite the high correlation between the average change in the share price of CEFs and ETFs, there is dissimilarity in their portfolio holdings. Of the 36 distinct companies in the top 10 holdings of CEFs in the sample, only 6 were common to ETFs.

Most surprising is the similarity in the holdings of the real estate ETFs. The adjacent table illustrates the most recent holdings of the real estate ETFs in our sample. Of the top ten holdings of the real estate ETFs, eight stocks were common to all. This has several implications—all other things being somewhat equal. The first is an investor would likely own the least expensive ETF as there is no portfolio selection benefit. (This may account for the huge volume turnover for URE as it trades as a single digit stock.) The second being arbitrage potential among the real estate ETFs. (Note the comparable YTD declines for 4 of the 5 ETFs.)

Ponderings

The fact that CEFs average daily trading volume is so much less than that of ETFs leads to the question: Why? Is it because there is something in the closed-end fund regulatory structure that precludes significant trading? Is there a technical or organic reason why there is less institutional ownership? The fact that URE (an ETF) has only $57 million[2] in assets, 1.575 million shares outstanding and trades on average 977,000 shares daily should mean that RNP, a CEF, with $1.663 billion in assets and 48.3 million shares outstanding should be on average trading more than .5% of its outstanding shares. As we know, trading begets trading. So, it would appear on the surface CEFs have little support on Wall Street. This means the likelihood of greater potential for mispricing.

Disclosures: Actively trades URE; long IGR and RNP

---

[1] Average daily trading volume varies by source due to calculation period and time frame.

[2] I used ETFConnect as my source; other data services have this number around $150 million.

Comments

[Scott F:]

I don't know; this is pretty thin gruel. ETFs and CEFs are not highly comparable because of CEFs' extensive use of leverage and ludicrous management fees. CEfs should be heavily discounted, considering their poor performance and unresponsive management (have any of you listened to one of their conference calls?). When you factor in their suspension of dividend payments for not meeting asset requirements, you really wonder if the author has done his homework. Lastly, what is the point of an article on arbitrage when you cannot borrow the shares? This does seem like random thoughts with unreadable graphs versus any concerted thinking.

2008 Dec 14 10:48 AM

[Alan Young:]

I agree with Scott that the article lacks usefulness, but for a different reason. REIT CEFs typically pay out 15-30% yield, vs. 5-8% (?) for ETFs. So comparing NAV charts doesn't tell you anything; you have to compare total return.

2008 Dec 14 02:42 PM

[Joe Eqcome:]

The most disappointing thing about this comment is that the person didn’t read the article or the conclusions carefully. The article suggests that a strategy of CEF and ETF arbitrage don’t make sense for all the reasons stated in the article. I actually put it in bold letters so it wouldn’t go undetected. For your benefit I’ll restate it again below:

“Due to the similarity of ETF portfolio holdings, there may be greater arbitrage potential among the real estate ETFs than between ETFs and CEFs. Clearly, given the relative trading volumes, ETFs are a superior trading tool to CEFs for real estate.”

Beyond that, the comments were factually inaccurate and off the mark. Let me deconstruct some of these comments.

1) “CEFs and ETFs and not comparable because of leverage.” There are many ETFs that employ leverage and they go by such names such as “ultra” “2x” and “double”. You can buy them either on the long or short side. In fact URE, a CEF that’s included in my study, is an ETF that employs leverage. So, this is a factually incorrect statement.

2) “CEFs should be heavily discounted given their poor performance and unresponsive management” Anyone who has followed CEFs recognizes that the best time to buy CEFs is when they’re at deep discounts, inexpensive and truly reviled. (You might be a good leading indicator for their recovery.) To dismiss them for that reason is like suggesting we should buy them when they’ve performed well and are at premium? History will demonstrate this is a flawed strategy.

3) “What’s the point of an article on arbitrage when you can not borrow the shares?” While you may be right that it’s difficult to short CEFs in term of volume, one certainly could go long a CEF and short an ETF.

I would hope that in the future that you read other peoples’ work more carefully and support your criticism with facts rather than conjecture and suppositions. The investment world is not a world that supports proof by assertion.

Joe Eqcome



On Dec 14 10:48 AM Scott F wrote:

> I don't know; this is pretty thin gruel. ETFs and CEFs are not highly
> comparable because of CEFs' extensive use of leverage and ludicrous
> management fees. CEfs should be heavily discounted, considering their
> poor performance and unresponsive management (have any of you listened
> to one of their conference calls?). When you factor in their suspension
> of dividend payments for not meeting asset requirements, you really
> wonder if the author has done his homework. Lastly, what is the point
> of an article on arbitrage when you cannot borrow the shares? This
> does seem like random thoughts with unreadable graphs versus any
> concerted thinking.

2008 Dec 14 06:27 PM

[Joe Eqcome:]

Scott F

There's an article in today's WSJ (12/15/08) regarding leveraged ETFs. It entitled, "Are ETFs Driving Late-Day Turns?
Leveraged Vehicles Seen Magnifying Other Bets; Last-Hour Volume"
Hopefully, this will help expand your knowledge base regarding leveraged ETFs.

Joe Eqcome

2008 Dec 15 12:56 PM

[Gruber:]

Judging from Scott F's comment, it seems like he couldn’t tell the difference between an ETF and a parked car let alone an ETF and a closed end fund. His limited understanding of the arbitrage compounds his problem. If he really wants to make a contribution he should do it by conserving electrons.

2008 Dec 18 12:14 AM