One of the smaller, but still very interesting types of securities are closed-end funds (CEF).  While traditional mutual funds (technically called open-ended funds) are much larger, the CEF market dates back to 1893 – 30 years before the first mutual fund.

The major distinction between the two fund types (ignoring exchange-traded funds, or ETFs) is that mutual funds continuously create new shares as new investors come into the fund and shares are created or redeemed after business hours at the net asset value (NAV) directly with the fund company.

In contrast, CEFs are closed to new capital after they begin operating, and since there are no new shares created or existing shares redeemed, a fixed number of shares trade throughout the day among investors rather than directly with the fund company.

What makes CEFs so interesting is that the market prices deviate, sometimes significantly, from the NAV of the fund, which is known to all investors.  In the market, there is almost always a premium or discount to the NAV of the fund.

This fact flies in the face of the Efficient Market Hypothesis (EMH) because the current price of a security should reflect all available information about the security – especially if the value is known.

The fact that the premiums and discounts persist is known in academia as the ‘closed end puzzle.’  In theory, investors would see a sizable discount, buy enough shares to have voting majority and then liquidate the fund at a profit.

In fact, this does happen sometimes, as it did in 2005 when Harvard’s endowment took a large stake in the Korea Equity fund and then encouraged other shareholders to liquidate the fund, but it’s not very common.

Even more puzzling than funds trading at a discount are funds that trade at a premium: why would investors be willing to pay 35 percent more for a portfolio than its worth, as they are currently for the PIMCO Global StocksPlus & Income fund (PGP).

There are some technical reasons that explain some of the discounts and premiums, but not enough to explain them entirely.  Some have hypothesized that the discounts reflect the present value of the fees, which are usually quite high, but that doesn’t explain the funds that trade at premiums.

What I find interesting about CEFs is that the premiums and discounts offer a little bit of a signal about investor sentiment.

When market prices are substantially above the NAV, it can be a signal that too many investors are chasing the same idea, are probably overpaying and isn’t a good value right now.

Alternatively, if market prices are substantially below their NAV, it may be a signal that the asset class is too out of favor right now with investors and if you’re a contrarian, it may be a good buy.

But it’s not enough to look at the premiums and discounts themselves, you have to look at the premium or discount compared to what it has been in recent years.  If a CEF ‘always’ has a 10 percent discount, then the discount doesn’t mean much.  If, however, the discount goes to 20 or 30 percent, it may signal a bargain.

Right now, taxable bonds (which basically means all bonds other than municipal bonds) are trading at substantial discounts.

For example, one fund that trades inflation-protected bonds has had an average discount of -12.76 percent over the last three years.  Right now that discount is -16.47 percent.  That might not sound like a big deal, but the current discount is three standard deviations away from the average discount, which is actually huge.

So, why aren’t we going to pounce all over it?  We have an allocation to inflation-protected bonds after all.

Even though it’s cheap in a relative sense, it has low credit quality relative to our inflation protected exposure (BBB- for this fund versus AAA for our fund), has high fees (0.95 vs. 0.20 percent), uses a lot of leverage (one-third of every dollar in the fund is borrowed), has experienced higher volatility than our choice and has had lower returns.

Indeed, the cheapness of this particular fund hardly seems worth it to us when you put it into context, but what’s interesting is that these characteristics have been true for a while and don’t completely explain the drop in price.

There could be some information that we don’t have, which is partly why we don’t trade discounted closed-end funds, but it’s a good, albeit loose, signal that taxable bonds are out of favor right now, which, to me is a good sign.