Weekly Fund Spotlight: Market-Beating Value In Preferred Stock CEFs

Summary

• Preferred stock CEFs have had a tough 2018 so far.
• Market-beating value can be found in preferred stock CEFs.
• My top pick from the sector is FFC.

2018 has not been a good year for preferred stock CEFs so far. The removal of several preferred stock funds from the ISE High income Index [YLDA], tracked by YieldShares High Income ETF (YYY) and ETRACS Monthly Pay 2xLeveraged Closed-End Fund ETN (CEFL), sent prices reeling at the start of January (see "CEFL/YYY Rebalancing: Duds And Opportunities"). Last week, distribution cuts from Flaherty & Crumrine funds caused more prices to tumble (see "Weekly CEF Roundup: CEFs Rebound While Distribution Cuts Opening Opportunities").

Given this opportunity in preferred stock CEFs opening up, I wanted to take a Spotlight of this class of CEFs. The searchable tag for the Weekly Fund Spotlight is "fundan."

According to CEFConnect, there are 16 funds in the "preferred stock" category in the database. Basic details about the funds are shown in the table below. I also included the benchmark ETF, the iShares U.S. Preferred Stock ETF (PFF), in the last row of this table.

Ticker	Fund	Yield	Discount	z-score	AUM/$m	Leverage	Baseline expense	Active expense
(DFP)	Flah & Crum Dynamic Pref & Inc	7.32%	-6.81%	-3.6	499	32.52%	0.99%	0.28%
(FFC)	Flah&Crum Pref Secs	7.43%	-8.81%	-4.4	890	32.81%	0.81%	0.14%
(FLC)	Flah&Crum Tot Return	7.21%	-8.41%	-4.2	213	32.86%	1.22%	0.45%
(FPF)	First Trust Inter Dur Pref & I	7.85%	-6.82%	-4.1	1511	30.28%	1.31%	0.54%
(HPF)	JH Preferred Income II	8.35%	-3.91%	-3.5	448	33.83%	1.25%	0.46%
(HPS)	JH Preferred Income III	8.18%	-3.62%	-2.6	555	33.42%	1.23%	0.45%
(HPI)	JH Preferred Income	8.29%	-5.51%	-3.8	593	33.68%	1.24%	0.46%
(JPI)	Nuveen Preferred & Income Term	7.93%	-8.25%	-2.9	1100	27.60%	1.25%	0.51%
(JPC)	Nuveen Pref & Income Opps Fund	7.09%	-7.43%	-4.1	587	33.58%	1.21%	0.44%
(JPS)	Nuveen Pref & Inc Securities	7.86%	-9.10%	-5.1	2097	33.09%	1.26%	0.48%
(JPT)	Nuveen Pref and Inc 2022 Term	6.42%	-4.50%	-4.0	128	19.82%	0.66%	0.08%
(LDP)	Cohen & Steers Ltd Dur Prf&Inc	7.43%	-7.65%	-4.8	785	28.67%	1.15%	0.42%
(PDT)	JH Premium Dividend Fund	7.41%	5.56%	-0.1	721	33.12%	1.39%	0.57%
(PFD)	Flah&Crum Preferred Income	6.88%	-5.75%	-3.9	159	32.77%	1.43%	0.61%
(PFO)	Flah&Crum Pref Income Opps	7.29%	-8.52%	-4.0	148	32.72%	1.47%	0.64%
(PSF)	Cohen & Steers Select Pref&Inc	7.67%	-3.57%	-3.4	332	27.84%	1.15%	0.43%
(PFF)	iShares Preferred Stock ETF	5.58%	n/a	n/a	17440	0%	0.47%	0.00%

Click to enlarge

(Source: Stanford Chemist, CEFConnect)

Expense ratio

The chart below shows the expense ratios for the preferred stock CEFs. One metric that I have devised to account for both the leverage of a CEF as well as the expense that one would have had to pay for a similar passive strategy is the "active expense" metric. This metric normalizes the baseline expense (which excludes interest cost) to 0% leverage, then subtracts from that the expense ratio for the passive ETF. The active expense ratio is therefore the additional fee that one is paying for active management versus a passive approach. I chose the iShares Preferred Stock ETF as the passive benchmark. PFF charges 0.47% in management expense ratio.

We can see that the preferred stock funds charge between 0.66% (JPT) and 1.47% (PFO) in baseline expense, whereas the active expense ratio calculates to be between 0.08% (JPT) and 0.64% (PFO). The active expense ratios are quite reasonable in my opinion, given how much some of these funds have beaten the benchmark as we'll see below.

(Source: Stanford Chemist, CEFConnect)

Portfolio

I will not be looking at individual portfolios in this Spotlight. Instead, an important consideration is the leverage of each fund. Leverage is a double-edged sword because it increases both the potential reward and risk of a fund.

All of the preferred stock CEFs are leveraged, with leverage ratios of between 19.82% (JPT) and 33.83% (HPT).

(Source: Stanford Chemist, CEFConnect)

We also consider the market capitalization of each fund. Some investors may not feel comfortable with investing in CEFs that are too small in size.

As shown in the chart below, fund sizes range from $128 million (JPT) to $2097 million (JPS). The benchmark ETF, PFF, is a giant with AUM of about $14.4 billion, and was excluded from the chart as it would distort the rest of the chart.

(Source: Stanford Chemist, CEFConnect)

Finally, it should be noted that the preferred stock CEFs may not contain 100% preferred stock. The chart below indicates the % of the CEF's portfolio that is in preferreds. This ranges from 30.73% (FPF) to 100% (PFF).

(Source: Stanford Chemist, CEFConnect)

If a fund is not fully invested in preferreds, where is the rest of the allocation going? Let's take a look at the six CEFs with <90% preferred exposure one by one to see what else they hold.

• FPF has 30.73% in preferreds and 64.98% in "Investment Funds" per CEFConnect. There is no indication of what this means exactly on the fund website either, although looking at their latest holdings gives a clue. I surmised that FPF's 30.73% "preferreds" allocation includes $25, $1000 and $1 million par preferred securities, while the fund's 64.98% allocation to "Investment Funds" describes its allocation to capital preferred securities, which are instruments similar to preferreds but which also have some features of corporate bonds.
• PDT has 65.97% in preferreds and 33.33% in equities. This fact should be borne in mind for anyone investing in this fund.
• JPS has 76.49% in preferreds, 15.87% in "cash alternatives" and 7.63% in convertibles.
• JPC has 84.51% in preferreds, 9.96% in corporate bonds and 2.62% in equities.
• HPF has 89.45% in preferreds, 7.80% in equities and 1.94% in corporate bonds.
• HPI has 89.79% in preferreds, 7.62% in equities and 1.49% in corporate bonds.

Yield

The following chart shows the distribution yields on price for the preferreds CEFs. These vary from 6.42% (JPT) to 8.35% (HPF). For reference, the benchmark PFF has a trailing twelve months [ttm] yield of 5.58%. The coverage of the distribution (as calculated from CEFConnect data) is also included in the chart below.

(Source: Stanford Chemist, CEFConnect)

However, note that many preferreds CEFs have been forced to cut their distributions in recent months due to the potent combination of rising short rates, increasing leverage expenses, and declining long rates, leading to lower returns on reinvested capital.

For instance, Flaherty & Crumrine's largest fund, FFC, has cut their level of distributions 3 times in the last 13 months (the cut from $0.136 to $0.128 in December 2016 is just missed on the left of the graphic).

(Source: CEFConnect)

Therefore, given the above, having >100% coverage should not be viewed as an ironclad guarantee that a fund will be able to sustain its distribution. Like UNII, these figures are lagging because they are not able to incorporate the effect of higher future expenses and/or lower income flows.

Performance

The following charts consider the historical NAV performance of the funds. NAV performance rather than price performance is considered because the latter is highly dependent upon changes in the premium/discount values of the fund, and as such do not accurately reflect the investing acumen of the fund managers. Furthermore, note that the performance figures are not adjusted for leverage, as doing so involves introducing additional assumptions to the data. As the preferred stock CEFs have relatively similar leverage ratios, not accounting for leverage should not have a major impact on the conclusions of this section.

The NAV performance of the preferred stock CEFs over the past year has generally been strong, ranging from +4.16% (PDT) to +14.69% (JPS). Here we can see how PDT's 33% equities exposure, which on cursory inspection contains significant exposure to the underperforming utilities and MLP sectors, have hurt performance over the past year. PDT was the only CEF that underperformed the benchmark ETF, PFF (+4.6%).

(Source: Stanford Chemist, CEFConnect)

On a 3-year basis, annualized NAV returns ranged from 6.15% (PDT) to 9.78% (DFP). Incredibly, all 15 CEFs managed to beat the benchmark ETF, PFF (+3.72%). This conclusion holds even if leverage is accounted for, as the performance gap between PFF and the worst performing CEF, PDT, is very wide. Note that JPT could not be included in the chart as it was only incepted last year.

(Source: Stanford Chemist, CEFConnect)

13 preferred stock CEFs have 5-year NAV histories. Again, all of them beat the benchmark ETF (+4.7%), with annualized NAV performances ranging from +7.28% (HPF) to 9.70% (FFC).

(Source: Stanford Chemist, CEFConnect)

At this juncture, I wanted to compare the preferred stock investing acumen of the various fund houses. To do this, I simply averaged the NAV performances of all of the funds under that family for each time period.

The data below shows that Flaherty & Crumrine funds have generally performed the best on a NAV basis over long time scales (5 and 10 years), while John Hancock funds have done the worst. On shorter time scales (1 and 3 years), Cohen & Steers and First Trust are competitive, even slightly edging out F&C.

Fund house	1-year NAV	3-year NAV	5-year NAV	10-year NAV
Flaherty & Crumrine (DFP, PFO, FFC, PFD, FLC)	13.24	9.13	9.34	10.83
John Hancock (HPI, HPF, HPS, PDT)	11.53	6.63	7.77	8.91
Nuveen (JPS, JPC, JPT, JPI)	10.54	9.16	8.71	6.69
Cohen & Steers (LDP, PSF)	13.63	9.26	9.00	-
First Trust (PFP)	13.42	9.55	-	-
Benchmark PFF	4.63	3.72	4.71	5.03

Click to enlarge

(Source: Stanford Chemist, CEFConnect)

Valuation

Now we get to my favorite section. The following chart shows the premium/discount values for all of the preferreds CEFs. We can see that with the exception of PDT (+5.6% premium), all of the preferred stock CEFs are trading at discounts, ranging from -9.1% for JPS to -3.6% to PSF. PDT is clearly an outlier here, I'd recommend selling it if you own it and replacing it with one of the other preferred stock CEFs in the list.

(Source: Stanford Chemist, CEFConnect)

I'm going to do a Douglas Albo favorite here and compare the 1-year price return of the preferred stock CEFs with their 1-year NAV returns. The funds are arranged in order of increasing difference between price and NAV return. In other words, funds whose NAV returns have most outpaced their price returns are shown on the far left of the chart.

(Source: Stanford Chemist, CEFConnect)

We can see from the chart above that for many of the preferred stock CEFs, NAV returns have far outpaced price returns, with PFO having the most striking difference (-14.45%). PDT is again an outlier, being the only fund showing the reverse behavior, with price outpacing NAV by +4.25% over the past year. This suggests that there is tremendous relative value in some of these preferred stock CEFs.

The 1-year z-scores of the preferred stock CEFs are shown in the chart below. The numbers suggest major carnage in this sector, with 13 out of 16 CEFs showing z-scores of under -3, which represents extreme relative undervaluation and presents ideal conditions for a short-term bounce.

(Source: Stanford Chemist, CEFConnect)

However, while these z-scores are extreme, note that reversion to the mean is not always guaranteed. It may well be possible that instead of discount reverting to the mean, the mean moves towards the discount, which I guess still counts as reversion but not in a manner that benefits the investor.

Verdict

(Normally exclusive to members of the Cambridge Income Laboratory, but is released to public as part of our free trial promotion)

With so much damage being inflicted on preferred stock CEFs recently, one might be surprised to find that the benchmark ETF, the iShares U.S. Preferred Stock ETF (PFF), is down only -1.26% year to date. Therefore, it seems that investors are selling out of CEFs in anticipation of trouble in preferreds ahead, even if the underlying assets themselves are holding steady.

PFF Total Return Price data by YCharts

Why are investors selling out of preferred stock funds?* One reason may be fear of rising rates. Preferreds are typically regarded as fixed income instruments, with long duration, and might be expected to perform poorly in a rising rate environment in which we are now. However, I offer the following arguments against this rather simplistic view:

1. Many preferreds issued after the Great Recession have floating or fixed-to-floating features. While admittedly, many of these may become called before they actually become floating, this does help to keep the price of the preferreds close to par, limiting duration risk. Moreover, assuming a rising rate environment, reinvestment risk is also not an issue.
2. When "everyone" expects something to happen, the market has a habit of behaving in the opposite way. Might this apply to rising yields? Bond vigilantes have been calling for the death of the bond bull market for quite some years now. While I'm no macro analyst, it seems to me that economic indicators are firing on all cylinders and sentiment is very optimistic. An unexpected disappointment in growth either domestically or internationally may throw a major wrench in the economic engine.
3. Related to point 2 above, one must remember that markets are forward-looking, and securities are priced accordingly. Could it be that the effect of rising rates has already been priced into preferred stock CEFs? In that case, one might conclude that most of the damage in rising rates has already been inflicted, giving us an ideal opportunity to buy low into depressed assets.

(*It should be pointed out at this juncture that leveraged CEFs, not benchmark ETFs, have been hurt by rising short rates leading to increased leverage expenses. This could be a contributing factor to the underperformance of preferreds CEFs versus ETFs.)

Moreover, preferred stocks offer excellent diversification to traditional stock and bond portfolios. As shown in the correlation matrix below, preferred stock only has 0.32 correlation investment grade bonds (LQD), 0.11 correlation with long-term treasuries (TLT), 0.57 correlation with high-yield bonds (JNK) and 0.46 correlation with U.S. equities (SPY) over the past 5 years. Instead, the highest correlations are between investment grade bonds and treasuries (0.84), and between high-yield bonds and stocks (0.66).

	PFF	LQD	TLT	JNK	SPY
PFF	1	0.32	0.11	0.57	0.46
LQD	-	1	0.84	0.19	-0.07
TLT	-	-	1	-0.12	-0.35
JNK	-	-	-	1	0.66
SPY	-	-	-	-	1

Click to enlarge

(Source: InvestSpy)

Volatility metrics are also very favorable for preferred stocks, with PFF having amongst the low volatility, beta, daily variation and maximum drawdown out of the major asset classes over the past 5 years.

Ticker	Annualized Volatility	Beta	Daily VaR (99%)	Max Drawdown	Total Return
PFF	5.3%	0.20	0.8%	-8.7%	25.4%
LQD	5.2%	-0.03	0.8%	-8.6%	19.0%
TLT	12.5%	-0.37	1.8%	-17.9%	19.9%
JNK	6.1%	0.34	0.9%	-16.9%	20.5%
SPY	11.8%	1	1.7%	-13.0%	108.8%

Click to enlarge

(Source: InvestSpy)

Hopefully, the above discussion together with the data presented above make a reasonable case for why I think preferred stock CEFs deserve a long and hard look either for tactical trading or for inclusion in an income portfolio. Not only do they represent significant value now (wide discounts and low z-scores), the vast majority of them have also managed to beat the benchmark ETF, PFF, over multiple time frames.

We don't have any preferred stock exposure in either the Cambridge/CIL100 or MIN portfolios right now, and I will be looking to initiate in this sector in the near future. Which fund do I like best?

My pick of a preferred stock CEF is the Flaherty & Crumrine Preferred Securities Income Fund (FFC) for the following reasons:

• Its discount of -8.81% is the 2nd-widest out of the peer group. (Update: I've realized that the discount is probably closer to -8%. CEFConnect has updated FFC's price but not its NAV, giving the illusion of a wider discount than actually is.)
• It has an extremely low 1-year z-score of -4.4, which is the 3rd-lowest of the group.
• Its NAV returns over the past year have outpaced price returns by 12.75%, indicating significant temporary undervaluation.
• Flaherty & Crumrine specializes in preferred securities, and its funds have shown the strongest historical NAV performances over the long-term compared to the other fund houses.
• FFC is ranked in the upper-third of all funds in the peer group for 1, 3 and 5-year NAV total returns. Notably, for 1-year returns it is ranked 3rd while for 5-year returns it is ranked 1st.
• Its baseline expense ratio of 0.81% is the 2nd-lowest of the peer group. (The fund website gives 0.89% as the baseline expense), and the active expense is only 0.14%. In other words, you're only paying 14 bps for peer group and benchmark beating performance.
• FFC has fixed-to-float exposure of 73%, which can act to limit NAV downside during rising rate environments.
• It yields a solid 7.43% (NAV yield 6.77%) with a 106% coverage ratio.

The top 25 issuers of FFC are shown in the table below, and include many recognizable financial names in the top holdings.

(Source: F&C)

The credit profile of FFC is shown below. The majority of the issues are in the crossover "sweet spot" in the BBB/BB region, where there is a nice balance of decent yield and relatively low default risk.

The geographical profile shows that FFC is mostly domestic focused, with about 75% of weighting in U.S. issues. This means that 25% of issues are from overseas.

(Oh, and in case anyone missed it above, anyone owning PDT is recommended to sell and replace with another preferred fund such as FFC.)

March 7 update: FFC has returned +3.80% since the idea above was released to members while PDT has returned -4.29%, good for a differential of over 8% only (equivalent to one year's worth of distributions from PDT!). Premium/discount mean reversion strikes again!

This article was originally published to members of the Cambridge Income Laboratory 1 month ago.

Analyst’s Disclosure: I am/we are long FFC, JPS. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Comments

[patrickhess]

FFC pricing recovered now. What is your current pick?

[Saint Mark]

Another terrific article. Thank you! I just entered a position in PFS. I’ve traded into it 3-4 times over the last decade when the price gets beaten down. Not sure it’s at the lows for this cycle but think it’s close enough that the risk/reward is worth it. The 8% yield looks good and it’s about in the middle of the CEF group in terms of leverage.

[bhef]

Thanks for more useful insight.

[Stanford Chemist]

Thanks bhef!

[IndyDoc1]

Nice review. Long FFC

[Stanford Chemist]

Thanks IndyDoc!

[Saz_from_EU]

Thanks for the article, SC.
The only things i miss here are comparison of duration and fixed/fixed-to-floating division which i think are critical in this environment. But otherwise, ton of useful info.

[Stanford Chemist]

Thank you Saz - very good suggestion for the future!

[etfman]

FPF has 86% in $1000 Par Preferred Stocks and 14% in $25 Par Preferred Stocks when you read their holdings at the First Trust site.

[Stanford Chemist]

Thanks for the info etfman - much appreciated! Sometimes CEFConnect has weird labels which is why I had to look up the annual reports for FPF - but I guess I should have just checked the website first ;-)

[Oil Can]

First, really appreciate the article on CEF's - preferred or otherwise.

Own several others in your list, but not FFC.

So the general idea is - the price is right, the dis has already been cut several times, the yield is good, what's not to like?

Some CEF's will not cut their distribution, no matter what (well almost). It's part of the "managed distribution policy". What's with the FFC cuts - i.e., are they using a different distribution policy than many other CEFs?

Thanks.

[Stanford Chemist]

Hi Oil Can, thank you and thanks for the question!

Most fixed income CEFs will have a managed distribution policy. When earnings go down they will either have to reduce their distributions to realign with earnings, or pay out more of the distributions from ROC/gains. The question as to "when" to cut and "how much" to cut will depend on management....

[katmandu100]

I wanted to buy one of the CEF preferred, but I am waitnig to see if the yield stablized.
The discount vs premium looks good. Thanks for the commentary Stanford Chemist.

[Stanford Chemist]

Thanks katmandu!

[NV_GARY]

SC-
Very good info.
You might be correct about FPF, but I have to wonder- from their site as of 1/31/18:

Security Percent
Fixed-to-Floating Rate Securities 78.85%
Fixed Rate Securities 15.16%
Floating Rate Securities 5.99%

Makes it hard to suspect that there might be some more esoteric form of preferreds.
G

[Stanford Chemist]

Thanks for sharing Gary!

[Ed Bickford, CPA]

Good article! I own PSF and also 10 individual preferred issues. PSF is one of the 10 CEF's which I also own. In the recent downturn the individual Preferred's are down less than PSF, less then the other CEF's as well as the REIT's I own. I think the leverage in the Preferred CEF's causes them to decline more in a period of rising rates than holding individual preferred's so I'm not sure I would want to invest a Preferred CEF right now. What do you think?

[Pacman13]

Mostly agree with you , I have seen essentially the same pattern but FPF and LDP have held up as good as some individual preferreds.

[Stanford Chemist]

Thanks for sharing Ed! Yes, you are correct that leverage will exacerbate movements in either direction - definitely something to keep in mind!

[bmayfield]

SC, still trying to due my due diligence, I am watching FFC closer, but the 3 cuts are a negative. Is there a way to extrapolate where the break even would be on total return given 3 cuts per year for the next three years? I assume your safety score from 1 - 10 for FFC might be greater than 5 ( you mentioned for JPS)?

[du4sloop]

SC
Thank you for another great article and I'm in total agreement with your conclusions. I'm a long time holder of FFC and JPC and will continue to hold them. For those waiting for a better entry point perhaps consider buying a small position now and adding over time or any future weakness. Earned dividends can often equal or at least shrink any price difference one might gain by waiting. That's another nice feature of buying a high dividend paying instrument.

[Stanford Chemist]

Thanks du4sloop!

[dmhogan]

How do all these compare to SPFF? Would it be better because it has lower fees than a CEF? And pays the same?

[NV_GARY]

dmhogan-
I was looking thru SPFF- can't find it right now, but it looks like their earnings coverage is ~50%. ROC varies, but their 8937 doc evens it out to each month being about 5%- the monthly 19As are quite a bit higher- but don't count at tax time,
G

[Stanford Chemist]

Hi dmhogan, thanks for the question. ETFs have lower fees than CEFs but you lose the features of active management, cheap leverage and the ability to take advantage of premium/discounts. All of these can be "double-edged" swords so it does come down to your personal preference and investor profile as well

[bmayfield]

SC,

Great article. I thought you might have included UNII in the comparison. FFC has .0213 which only covers 18% of the distribution. You must have some gut feeling they might not reduce the distribution for awhile? Two others with better UNII are PDT (59%) and FPF (60%). But I suppose there is a possibility any of these CEFs could cut their distribution as you said.

[Stanford Chemist]

Hi bmayfield, thanks for the comment! As you can see from the article I used % coverage rather then UNII. Both UNII and coverage will be lagging metrics, but I prefer coverage because it's easier to visualize - telling us whether the earnings are covering the distribution or not regardless of the UNII bank. Moreover some funds with positive UNII cut, and others with negative UNII don't cut, so it's not very useful as a short term timing tool IMO. Finally, I'm actually quite agnostic on dividend cuts -- theoretically a lower dividend is made up with better NAV returns going forward!

[bmayfield]

It is probably right in front of me but I do not see the coverage # on any of the cef connect tabs.

[Stanford Chemist]

bmayfield, I take the earnings number from CEFConnect and divide by the per share distribution. You can also go to CEFdata where they do the coverage for you (e.g. http://bit.ly/2DdDF9L).

[alphabettr]

Oh, and thank you for the article SC. I already have a couple of preferred CEFs but FFC looks interesting.

[alphabettr]

Might be lack of experience on my part but given that we have an economy that could be overheating, full employment, rising inflation (possibly exacerbated by trade wars), possibility of wages rising (ie teacher's contract), the Fed saying 3 more increases this year, the 10 year yield creeping up, etc. - that there might be a even better entry point coming up for preferreds in a few months? Just curious what others might think. I may sit tight and look for better opportunities.

[Stanford Chemist]

HI alphebettr, thanks for sharing your insight!

[Pablo]

Typical. FFC is one of the few I don't own. Will check it out!

[Stanford Chemist]

Thanks Pablo!

[jradwan]

Thanks for this extensive report. Just an FYI and since I own JPS I was curious when I noticed your 1-year Z-score at -5.1, cefconnect and cefdata has it as -1.90.

[Stanford Chemist]

Hi jradwan, thanks for the comment! This article was first released to members of Cambridge Income Laboratory 1 month ago (as noted in the article) so the data may have changed since then.

[Vivek Rao]

I think your reports for non-subscribers are useful even though they are delayed, but I wish that before each data table you would list the "as of" date, so that the vintage would be clear to readers.

[Stanford Chemist]

Hi Vivek, thanks for the great suggestion!