Benjamin Graham et al. on Preferred Stocks

A query regarding Graham’s position on preferred stock came to my attention, so I looked it up:

Certain general observations should be made here on the subject of preferred stocks. Really good preferred stocks can and do exist, but they are good in spite of their investment form, which is an inherently bad one. The typical preferred shareholder is dependent for his safety on the ability and desire of the company to pay dividends on its common stock. Once the common dividends are omitted, or even in danger, his own position becomes precarious, for the directors are under no obligation to continue paying him unless they also pay on the common. On the other hand, the typical preferred stock carries no share in the company’s profits beyond the fixed dividend rate. Thus the preferred holder lacks both the legal claim of the bondholder (or creditor) and the profit possibilities of a common shareholder (or partner).

These weaknesses in the legal position of preferred stocks tend to come to the fore recurrently in periods of depression. Only a small percentage of all preferred issues are so strongly entrenched as to maintain an unquestioned investment status through all vicissitudes.

Experience teaches that the time to buy preferred stocks is when their price is unduly depressed by temporary adversity. (At such times they may be well suited to the aggressive investor but too unconventional for the defensive investor.)

In other words, they should be bought on a bargain basis or not at all.

Another peculiarity in the general position of preferred stocks deserves mention. They have a much better tax status for corporation buyers than for individual investors. Corporations pay income tax on only 15% of the income they receive in dividends, but on the full amount of their ordinary interest income. Since the 1972 corporate rate is 48%, this means that $100 received as preferred-stock dividends is taxed only $7.20, whereas $100 received as bond interest is taxed $48. On the other hand, individual investors pay exactly the same tax on preferred-stock investments as on bond interest, except for a recent minor exemption. Thus, in strict logic, all investment-grade preferred stocks should be bought by corporations, just as all tax-exempt bonds should be bought by investors who pay income tax.

In the last paragraph, Mr. Graham recognizes that tax differences are important – very important! Using his figures the equivalency ratio within a corporation is a stunning 1.9x – in other words, it took $1.90 in interest to provide the same after tax income as $1.00 in dividends. For an individual, the equivalency ratio was 1:1.

It seems quite clear to me that under these conditions there will be very little left on the table for individual investors – Mr. Graham’s target audience – after corporations have picked through the offerings.

Other than this, the passage is sorely lacking in numeric analysis and opinion based on specific fact. Mr. Graham acknowledges that there is some price – some yield – at which a preferred share becomes superior to a given bond, but does not provide any analytical framework that will allow an interested reader to determine how that price – that yield – might be determined.

The market has changed dramatically since the early ’70’s. Besides the taxation differences between US-then and Canada-now already noted, there are regulatory elements in play that make preferred shares an attractive way to raise capital for banks and some utilities.

However, my main objection to the passage is that it is too rigidly doctrinaire. The world as presented is black and white, with safety on one side and profit on the other. In fact, the real world contains many shades of grey, which are ignored.

The revised edition cited contains further commentary by Jason Zweig:

Preferred shares are a worst-of-both-worlds investment. They are less secure than bonds, since they have only a secondary claim on a company’s assets if it goes bankrupt. And they offer less profit potential than common stocks do, since companies typically “call” (or forcibly buy back) their preferred shares when interest rates drop or their credit rating improves. Unlike the interest payments on most of its bonds, an issuing company cannot deduct preferred dividend payments from its corporate tax bill. Ask yourself: If this company is healthy enough to deserve my investment, why is it paying a fat dividend on its preferred stock instead of issuing bonds and getting a tax break? The likely answer is that the company is not healthy, the market for its bonds is glutted, and you should approach its preferred shares as you would approach an unrefrigerated dead fish.

For all his colourful language, Mr. Zweig shows lamentable ignorance of bank regulation; this is perhaps partly due to his exclusive focus on the American market, in which dividends are taxable to an investor at the same rate as interest; hence the market is much less vibrant in the US than in Canada.

I will certainly agree with Mr. Zweig’s emphasis on the undesirability of call features – but a call is simply another element of investment risk, to be calculated and incorporated when determining the value of an asset.

I will note that according to his biography, Mr. Zweig is a journalist, not an analyst or portfolio manager. It is therefore not possible to gauge the value of his opinions by reference to his results.

5 Responses to “Benjamin Graham et al. on Preferred Stocks”

  1. adrian2 says:

    “this is perhaps partly due to his exclusive focus on the American market, in which dividends are taxable to an investor at the same rate as interest”

    That was once true, and may become true in the future; currently there is a significant tax advantage for a US individual to receive dividends vs. interest.

    BTW, in the last little while, even before the Canadian preferreds took off like a rocket, the US bank prefs I’ve been watching (Bank of America, Citi plus an ETF for an RBS pref) have significantly outperform the matching common equities.

    Adrian

  2. lystgl says:

    “Once the common dividends are omitted, or even in danger, his own position becomes precarious, for the directors are under no obligation to continue paying him unless they also pay on the common.”

    I don’t believe the gentleman has this right either. They are under obligation to pay whether or not the common share dividend is deleted altogether are they not? (cumulative dividend at least though they can suspend temporarily but must play catch up before paying the common?)

  3. jiHymas says:

    That was once true, and may become true in the future; currently there is a significant tax advantage for a US individual to receive dividends vs. interest.

    Well … OK. That’s the famous Bush Tax Cuts.

    Despite this, I’m happy with the overall sense of the paragraph – it’s very hard to invest in perpetuals when one of the two parties is stridently promising to eliminate the tax advantage at the next opportunity!

    The closest equivalent to the Canadian preferred market in the US is the municipals market, where an investor pays zero tax.

  4. jiHymas says:

    They are under obligation to pay whether or not the common share dividend is deleted altogether are they not? (cumulative dividend at least though they can suspend temporarily but must play catch up before paying the common?)

    That’s right – once the common dividend has been eliminated, any “adjustments” to the preferred dividend are an entirely separate decision. The two current preferred defaulters in Canada, IQW and NTL, both paid preferred dividends long after the common dividend had been eliminated.

    Graham’s phrasing is a little clumsy, I think, but he means that if the common dividend is paid, the company is obliged to pay preferreds; but there is no obligation to pay preferreds if the common dividend is eliminated.

    And yes, if the dividend is cumulative, the company has to catch up before the common can recommence. This became an issue in the GT.PR.A default.

  5. […] I commented on an extract from this book dealing with preferred shares in my early 2009 post, Benjamin Graham et al. on Preferred Stocks. Briefly, Mr. Graham was writing in another time, under a very different tax regime; I agree that […]

Leave a Reply

You must be logged in to post a comment.