Do Jubilee shares make any sense?

By Felix Salmon
April 16, 2012
" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

One of the more intriguing concepts to come out of the INET conference was Steve Keen’s idea for what he calls “Jubilee shares”. It’s not exactly new — he’s been writing about the concept since October 2010 — but he refined the concept for INET, and it has a bizarre kernel of genius to it, for all its flaws.

Here’s how it works. Right now, shares issued by a company represent the permanent equity capital of that company. If a company raises new equity capital, then the people buying that stock will have an ownership interest in that company in perpetuity. Under Keen’s proposal, none of that changes. But when those shares get sold, things start getting interesting.

At companies like Google, one class of shares automatically converts to another class when they’re sold. In Keen’s world, all companies would be a bit like Google. Not in terms of voting rights: each share would still carry the same voting weight. But there would be different share classes, all the same. Eight of them, to be precise.

As a rule, when companies issue Jubilee shares, they issue Class A shares — the highest class. And the way that Jubilee shares work, Class A shares would automatically convert to Class B shares when they were sold.

Now that wouldn’t be much of a change. Class B shares have all the same ownership and voting rights of Class A shares; the only difference between Class A shares and Class B shares is that when Class A shares are sold they become Class B shares, while Class B shares convert to Class C shares when they’re sold.

You can guess what Class C shares are like: they’re exactly the same as Class A shares and Class B shares, except that they convert to Class D shares when they’re sold. And so on and so forth, until you reach Class G shares. They convert to Class H shares when they’re sold, and Class H shares are actually very different indeed from all the others. Because Class H shares are not permanent equity capital at all: instead, they expire, worthless, on their 50th birthday.

If you hold Class H shares, you can trade in and out of them as much as you like: there’s no Class I. And you get full dividend payments and voting rights. But you also hold a piece of paper with an expiry date. As far as the cashflows from Class H shares are concerned, it’s basically just 50 years’ worth of dividends, and that’s it. (Although, if the company is sold, you get full participation rights.)

For a company like Berkshire Hathaway, which has little prospect of being taken over and which doesn’t pay a dividend, Class H shares would be close to worthless. On the other hand, for a company which is in clear decline and which is probably going to fail or get taken over in the next decade or two, Class H shares — at least the ones still far from expiry — would trade at only a very modest discount to Class A.

For companies going public, issuing Jubilee shares would be quite attractive, in some ways. The founders of Google and Facebook would feel much less need to give themselves super-voting rights, or to worry that IPO allocations are silly because they just end up getting flipped on day one, because the structure of the Jubilee shares would encourage shareholders to act like long-term owners rather than short-term traders.

For investors, Jubilee shares would also be attractive. Any company with Jubilee shares would have a very low stock-price correlation with the market as a whole — and investors like low correlations nearly as much as they like liquidity. And besides, Jubilee shares would be cheaper than normal shares, and it’s always nice to be able to buy equity in a company at a discount.

As for traders, Jubilee shares would be a very mixed bag. On the one hand, volumes would plunge. But on the other hand, bid-offer spreads would rise, and there would be a lot more opportunity to generate alpha and outperform the market by smartly navigating the various classes of stock.

Jubilee shares would work like a financial-transactions tax: on a mark-to-market basis, you’d take a loss every time you bought a stock. Shares would trade in seven main classes: A/B, B/C, C/D, and so on, with the first letter representing what the seller is selling, and the second letter representing what the buyer is buying. Since each class would trade at a lower price than the one before, the cost of doing a round-trip trade — of buying a stock and then selling it immediately — would be substantial. And I haven’t really thought through what might need to be done in the area of shorting and securities lending.

Still, there would surely be active trading, and the biggest profit opportunities, in Class H shares. Those shares would be highly specialized financial instruments, not least because they wouldn’t be fungible: each one would have a unique expiry date, and would be priced accordingly. Broker-dealers would trade them on the OTC market — it would be incredibly difficult to trade them on an exchange — and would tempt merger arbs and anybody else in the special-situations space with the promise of enormous profits. Dividend-related announcements would take on huge market importance, much more than they do now, and the difference between dividends and stock buybacks would go from being negligible to being enormous.

For all the active trading in such instruments, however, what you would not get would be a speculative bubble. The price of Class H shares would always be capped at the price of Class A/B shares, and Class A/B shares would be almost impossible to speculate in because volumes in that market would perforce be extremely low.

And so I think that Jubilee shares would indeed achieve what Keen intends them to achieve — the end of stock-market bubbles. They would also make investing in the stock market extremely difficult — something which can probably be considered a feature rather than a bug. Most investors would be forced to do their homework and really understand what they were buying; you wouldn’t get people logging on to E-Trade and buying thousands of dollars of a stock just because of something they saw on CNBC. And the huge current volume in ETFs — most of which is accounted for by speculative day-traders — would disappear overnight.

There are two ways that Jubilee shares might be introduced, neither of which is going to happen. One option would be for them to simply be imposed on the market by legislative fiat; that seems to be what Keen has in mind. That wouldn’t just be bad politics, it would be bad policy, since stock-market bubbles aren’t actually all that much of a problem. As we saw in 2000, they can wipe out enormous amounts of wealth when they burst, with surprisingly modest macroeconomic consequences, thanks to the fact that most stock-market investments are unlevered.

More to the point, legislating Jubilee shares would only make debt even more attractive than equity, as a funding source for companies — and that’s exactly what we don’t want to achieve. Unless and until Congress first abolished the tax-deductibility of corporate interest payments, the introduction of Jubilee shares would cause more harm than good in the markets as a whole, giving companies even more incentive to borrow money rather than to fund themselves with equity.

There is another way for Jubilee shares to arrive, however, and that’s for companies to issue them voluntarily. As far as I can tell, there’s no reason why a company couldn’t issue Jubilee shares rather than common stock tomorrow, were it so inclined. The market capitalization of any such company would certainly suffer: a founder wanting to maximize the mark-to-market valuation of her own stake in a company would never opt for such a structure. But for CEOs who prefer long-term control to paper wealth, Jubilee shares could be an attractive alternative to the current modish option — dual classes of shares with the founders’ shares having many more votes than everybody else’s.

But founders don’t need to issue Jubilee shares, now, thanks to the passage of the JOBS act. It’s now much easier for founders to retain control: with the 500-shareholder rule no longer in effect, founders can simply elect to stay private indefinitely, with a right of first refusal on any share sales and essentially complete control over where, how, and even whether their stock is traded. Given the attractions of private markets and the new powerlessness of the SEC when it comes to requiring companies to go public, it seems like Jubilee shares are to a large degree a solution to a problem which no longer exists.

Still, I’d love to see just one company try them out, if only to see what happens. Existing corporate stock can remain untouched, but new shares, be they sold directly to the public or given out to employees or used to buy some other company, could still be issued in Jubilee form. It would be fascinating to see where and how they traded.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

“most stock-market investments are unlevered.”

Is there a fundamental economic reason for this? Just curious.

Posted by Britonomist | Report as abusive

“Is there a fundamental economic reason for this? Just curious.”

No. The reason is because of limits on the fraction of equity that can be used as the basis for a loan.

If we had the same rule in real estate there would be no bubble in debt.

It’s just a reflection of the rules we make.

Posted by BrPH | Report as abusive

What would be far more interesting would be jubilee bonds, don’t you think?

Posted by BrPH | Report as abusive

@BrPH, the reason for those limits is that equities (being quite volatile) would rapidly wipe out anybody leveraging them 30:1. Even 10:1 leverage would destroy you most years.

Jubilee shares sound like a gimmick to me. Investing sensibly is by its very nature complicated enough. Any gimmicks will make the system more complicated, less flexible, and (in my opinion) ultimately less robust.

Posted by TFF | Report as abusive

Okay I guess what I’m getting at is this, assuming these rules don’t change, do I have no reason to worry about stock market bubbles in terms of macroeconomic effect? Because that is very reassuring.

Also, is there any economic reason why we can’t have a similar rule for real estate?

Posted by Britonomist | Report as abusive

Stock market bubbles have comparatively limited immediate macroeconomic effect. Consider the 2001 recession in which we saw a massive drop in the stock market yet just a brief dip in GDP. Stock investments aren’t leveraged, they aren’t made with tomorrow’s lunch money, and established companies can get along just fine without tapping the stock market for new cash. (You do see fewer IPOs after a crash, but that just means that companies are staying private a little longer.)

That said, the volatility of the stock market frightens away many people who ought to be investing the bulk of their retirement savings in equities. You have to be able to stomach (or ignore) a 30% drop in the market, and many people simply can’t do that. So there is a long-term cost to these bubbles…

You can make whatever rules for real estate you like. If you demand a higher downpayment, then you will get a less volatile market but people will need to save for more years before they buy. If you accept a smaller downpayment, then people can buy earlier in their life story but the market is more vulnerable.

Once upon a time the standard downpayment was 20%. That helped keep prices affordable, and forced people to save a substantial sum before they could buy. If you can’t save a 20% downpayment in five years, there is a good chance that a purchase at that price would excessively stretch your finances. I would favor a return to those policies, but such a change would depress the housing market significantly (and thus this probably isn’t the best time to do it).

Posted by TFF | Report as abusive

Okay thanks TFF

Posted by Britonomist | Report as abusive

It must have been a pretty uninspiring INET, ‘eh FS?

Posted by MrRFox | Report as abusive

This seems like a fascinating experimental idea to combat the epidemic of short-termism that’s rampant in the economy

Posted by EconMaverick | Report as abusive

I’ll answer the question posed in the title with a resounding “No.” This idea is crazy for a host of reasons.

I fail to see why it would prevent bubbles – I don’t see how the prospect that a share could be worthless in 50 years would have stopped the tech bubble in the late 90′s. People happily buying shares at a multi-billion valuation for a company that had never made money weren’t doing so because discounting cash flows told them that the years 51 to infinity dividends would be worth piles of money. They did it because they thought these companies would quickly dominate some market and/or because they thought they could flip to a fool at a greater price.

It would have a devastating impact on liquidity – not only because each sale of a share makes it less valuable, but also because the common shares of a company are now a series of classes (more than 8, since a class H share is also defined by its expiration date) rather than 1 identical class. This illiquidity probably also makes bubbles more likely, not less. A shared feature of both the residential real estate bubble and newly-public tech stocks during the bubble was transactions involving only a small fraction of assets driving the valuation of a much larger, illiquid group of assets. (For housing, it’s obvious that the asset is relatively illiquid. For the tech stock bubble, newly public companies had a small float, due to insider/VC ownership and lockups, so the trades of a float of about 10% to 20% of total stock were setting the price.)

Valuation of any class of shares becomes extremely complicated, because the non-expiring shares become more valuable if a company doesn’t pay dividends, allows a large number of Class H shares to expire, and then pays dividends to the now smaller remaining group of shares.

Building on that valuation point, corporate governance would be even more of a mess than it is now, due to these fights on the timing of dividends. Boards currently have reasonably well-defined fiduciary obligations in theory, even if in practice they often don’t live up to them. In this new world, what is a board’s obligation to holders of Class H shares that expire in 2013 or 2014? Is it to pay a dividend this year, being equitable to all shareholders? Or is it to delay a dividend, which benefits the vast majority of shareholders who don’t own such Class H shares?

Then there’s the regulatory arbitrage we’d get. I’d obviously much rather own a mutual fund or ETF that owns shares than actual shares, assuming that I can still trade in or out of the fund/ETF.

So, if our goal is to bring some of the worst of the complexity and regulatory arbitrage of the pre-crisis structured finance market to the equity market, while also handing dramatic arbitrary power to corporate boards, then this proposal is a winner.

Posted by realist50 | Report as abusive

Wouldn’t this complicated capital structure be totally worthless, because it can be circumvented by using derivatives?

I would guess it would be possible to set up a SPV, which would buy A class shares at IPO and then issue CFDs or any other similar instrument that would allow trading in those shares without actually moving them to the balance sheet of other investors.

You can tell I used to work in an investment bank :)

Posted by Petras_Kudaras | Report as abusive

Beyond idiotic, in Steve Keen’s own words:
“There’s a twist to my proposal of course: it wouldn’t be a liability that was abolished but an asset, but the intent is to stop the liability of debt…”

(Hint: a liability/debt is someone else asset)

Posted by alea | Report as abusive

Well put, realist.

And Petras, I was just thinking of ways to circumvent it myself. But yours is far more elegant!

Posted by TFF | Report as abusive

Surely large tranches of shares would just trade through an SPV? Buying and selling shares in the SPV rather than the actual shares.

This is similar to the way many large properties are often sold (ie you buy the company with the sole asset of the property rather than the property), although there it is about getting around stamp duties etc.

This would be even more likely if, as you say, a lot of the trading would be OTC because of the lack of volume.


Posted by Quarrel | Report as abusive

realist50 says that this proposal would have “a devastating impact on liquidity”, which is just another way of saying that it would discourage trading of shares for trading’s sake, which I’m sure in Keen’s mind is a feature rather than a bug.

Posted by Strych09 | Report as abusive

Strych09, read the comments above. Without additional regulation it would mostly just slant the market to “sophisticated investors” who can and do find ways to invest in things even without buying them directly.

Have pity on the little guy!

Posted by TFF | Report as abusive

Too complex to be useful.

Posted by DavidMerkel | Report as abusive

I think you have missed the game theoretic implications. Take a company like Birkshire Hathaway, for example. You mentioned that with no dividends or possibility that the company will be sold, class H stocks would be worthless. But that means class G shares are also worthless, because you would only be able to sell them at whatever price someone is willing to buy H class shares, which is $0. That means that class F is worthless, and so one all the way to A. It is not enough for the class A through G stocks to be non-expiring, but they also have to be fully transferable, because the value of Birkshire Hathaway stocks really only in the limit–the value comes from the amount that liquidation of the company will yield shareholders, but if that liquidation won’t come in our lifetime, and we can’t transfer stocks without them becoming perishable, then none of the stocks have value. This means that the firm would bleed equity until it is forced to offer regular dividends, which in turn reduces the profitability of the firm.

This backward’s recursion principle would apply to all classes of stocks: a class H stock would be priced at the discounted sum of 50 years worth of dividends, and a class G stock would be worth that plus the discounted sum of dividends expected before selling it, and so on. But ultimately, all of these classes of stocks have value if and only if the firm pays regular dividends. At this point, we have to call into question whether these should be called “dividends” at all–since it is now an obligatory payment needed to maintain the company’s capital valuation, it should be called “interest” not dividends, and recorded as an operating expense, not profits.

My point is what you have described is just an incredibly complicated reformulation of a financial instrument already available to corporations: a bond. Essentially, Keen wants to turn stocks into bonds, so that a stock is really a debt issued by the company that has to be repaid in 50+some odd number of years with interest. We could simplify the whole thing if we eliminated all 8 classes of stocks and simply specify that bondholders have voting rights.

Posted by econ2 | Report as abusive