Here’s a chart that ought to get your attention. It shows the performance of SG’s new Quality Income index backtested to December 1989. The index of quality income stocks is in blue, and a world index of stocks is in red. The quality index produced annualised net total returns of 11.6% compared to 5.6% for shares and 8.7% for government bonds.

Quality Income was less volatile than shares in general and in its worst year lost 33% compared to 49% for the market.

If you could have invested in 1990, you would even have beaten bonds!

But it looks as though the strategy really stared motoring this century when circumstances have favoured income shares. Record low interest rates mean the alternatives, cash and bonds, are less attractive, and a treacherous stock market may have encouraged a flight to quality. In the US, professor Aswath Damodaran finds that in previous decades the performance of high income stocks matched low income stocks at best.

SG started from the premise that the biggest driver of returns from the stock market is dividends. I don’t understand why that matters, in fact it so mystifies me I’ve asked for an explanation. To beat the market you must pick better stocks, which means finding shares that are undervalued, at least to a fundamental investor. Where the extra value is, is a different question to where the aggregate market returns come from.

Nevertheless, SG contends income is a proxy for value because a dividend helps safeguard investors against empire building-managers who would otherwise waste the money if they could get the fevered hands on it. Managers generally destroy value, but the market doesn’t realise.

Dividend payers also tend to be mundane businesses, underrated by the market.

But buying the highest yielders, the companies that have paid the highest dividends in relation to their current price is not the best way to capture that value. That’s where quality comes in.

Dividend can be cut, generally when a company can no longer afford them. It’s common knowledge that very high dividend yields are a sign the dividend is unlikely to be paid in full:

Which probably explains why the highest yielders do not generally perform.

SG says the traditional method of weeding out companies at risk, ensuring the dividend is covered by earnings or cash flow, does not work  well. It plumps for balance sheet quality instead.

This is where things get a little complicated. It employs a Distance to Default measure developed by KMV and used by credit agencies along side the familiar F_Score (it requires an F_Score of 7 or more out of nine). As far as I’m aware, Distance to Default is not available from data sources used by private investors, and the formula is complicated. Stockopedia has implemented a version of the index for screening stocks that substitutes Altman’s Z Score for the Distance to Default.

The quality aspect of the Quality Income index provides it with a double-whammy, not only do strong and improving finances indicate a safe dividend, but good quality companies, with safe balance sheets, tend to be underestimated by the market too.

Such companies are modestly contrarian, so SG is not too worried it’s stoking a bubble:

…we think the historic outperformance of quality and dividend paying stocks has less to do with investment fashion, and more to do with our behavioral hard-wiring. To the extent that our index is trying to avoid overconfidence, and the human tendency to pay too much for lottery tickets remains, we think it should be more robust than other savings vehicles.