Cash is king

Princely returns

*From 1 January 2004 to 5 March 2014

(i) 45 components of FTSE 350 in both January 2004 and March 2014 that had net cash on their balance sheet at both times.

(ii) 206 'survivors' of FTSE 350 that were in the index in both January 2004 and March 2014. Excludes investment trusts, banks and insurance companies.

Yet in some ways the findings are odd. After all, what's really strange is why so many companies should continually have more cash than debt - even if they have any debt at all - when, in effect, the government pays them to borrow money. That's because the interest paid on servicing debt is tax-deductible. So interest paid means taxes saved. With the standard rate of corporation tax at 23 per cent in the UK, a company that borrows at, say, 6 per cent really pays just 4.6 per cent after capturing the tax benefit. Not just that, the taxes saved also have a quantifiable benefit for shareholders since, in effect, they produce a stream of income that can be valued much like an annuity.

With all this going for debt, the wonder is that so few companies 'gear up' (ie, raise more debt relative to their equity) and maximise the benefit. Sure, it's understandable that some companies should have no net debt from time to time. But why do so many companies consistently carry net cash? For example, at the time of writing, of the components of the FTSE 350 index, 106 companies carry no net debt. In addition, another 15 have a nominal level of gearing, where net debt is less than 10 per cent of equity employed.

This phenomenon is also alive and well in the US, where the rules on the tax deductibility of interest payments are much the same. In a huge study of US companies, two academics - Ilya Strebulaev of Stanford University and Baozhong Yang of Georgia State University - found that between 1962 and 2009 on average 32 per cent of publicly-quoted companies had no net debt and that 10 per cent had no debt at all. In addition, there was a 60 per cent chance - odds of six to four on - that any company that carried no net debt in one year would do the same in the following year.

As to the reasons why, the academics struggled to find an explanation, although they highlighted interesting features of cash-rich companies that provide food for thought. On the positive side, they found that cash-rich companies were more profitable than the average. Their vast study found that on average 'zero-leveraged' companies (ie, those with no net debt) generated profit margins of 12.4 per cent compared with 9.2 per cent for the universe of companies (both leveraged and debt-free) in the project. Zero-leveraged companies were also more likely to pay dividends and spend more on research and development than the average of the universe; and they were less likely to have a deficit in their pension schemes and could expect their shares to trade at a higher ratio of price-to-book value.

Yet holding net cash may bring its dark side, too. The US academics also found that cash-rich companies were more likely than average to have a chief executive with a high level of share ownership. That could be good if you assume that her (or much more likely) his incentives really were aligned with shareholders. In those circumstances, the boss may have incentives to behave like the archetypal 'owner-manager', who does best for the company's shareholders because he is nurturing his own capital as well as theirs.

But where the boss owning lots of shares might be bad would be if you feared he was motivated to boost the share price by focusing on short-term measures at the expense of long-term ones. And - really worrying - the academics found that cash-rich companies were associated with smaller - and less independent - boards of directors. That implies a propensity for the boss to run the company as his own little fiefdom where high levels of cash effectively protect him from the strictures of shareholders and the discipline of the financial markets.

Given all this, the wonder - and the pity - is that the academics did not address the first question that an investor would ask: are shares in cash-rich companies likely to be good investments or bad?

This is our attempt to answer that question and it brings the short answer: more good than bad, although it's not that clear cut. First of all, however, we must throw in the caveats.

The major caveat is that ours is just a small study, using the components of the FTSE 350 index between 2004 and 2014. From those, we have deducted companies whose shares were not in the index at both the start and the end points. In addition, we have removed shares in investment trusts, banks and insurance companies on the logic that their balance sheets are not suitable for this exercise.

So the sample distils down to just over 200 companies, whose balance sheets and share price returns are examined using several snapshots in the period 2004-14. Purists might say that's not enough to be statistically significant. Fair point, although a sample of 200 is not to be scoffed at. So off we go.

The most obvious way to test for a link between leverage and share price returns is with a regression model. That may reveal if there is a pattern where, say, returns rise as companies hold less debt and more cash. Alternatively, the opposite scenario might apply where, thanks to the tax benefit of carrying debt, investment returns would rise in line with leverage. A regression model - effectively a scattergram - could help. This is where dots on a chart juxtapose - in our case - levels of leverage along the 'x' (horizontal) axis and share price returns along the 'y' (vertical) axis. We are looking to see the extent to which inputs on the x axis, which plots the 'independent' variable, drive outcomes on the y axis, which plots the 'dependent' variable. The aim is to find the 'line of best fit' between all the dots (it's the line that minimises the space between all of the dots and the line itself), to see which way that trend line points and to assess the extent to which the independent variable drives the dependent one.

Miserably, the line of best fit points nowhere in particular and this is quantified in the table, 'Phoney indicators'. This examines the relationship between falling levels of leverage and share price returns for 205 FTSE 350 companies over periods of one year, three years and five years starting from January 2004. The figures under the 'beta' column show the correlation between leverage and stock returns (in order to use a consistent measure for leverage, each company's net debt or net cash is expressed as a percentage of its stock market value in January 2004). If there were a perfect correlation - ie, returns rise by one unit for every unit that leverage falls and then gives way to net cash - the beta value would be 1.0. In general, the higher the beta value, the more that falling leverage and rising cash levels drive rising stock returns. The more that beta is a negative value, the more the opposite applies - that rising leverage drives better investment returns.

Phoney indicators

Relation between falling leverage and share price returns

See text for explanation

As it is - whether it's for returns on one year, three years or five years - leverage seems to have little influence on stock returns on average. Each value for beta is close to zero. True, it's interesting that the beta values for one year and three years are mildly negative, implying that in the years 2004 to 2006 companies were rewarded for carrying high levels of debt. That would be consistent with the spirit of recovery that followed the bursting of the 'dot-com' bubble in 2000-02 coupled with low interest rates and low inflation. However, over the five years 2004-08 beta becomes positive, implying that on average low debt/net cash brought better investment returns. That would be consistent with the shuddering reappraisal of debt's dangers that began with the collapse of Northern Rock in 2007.

But we should not take these figures very seriously. Or, at least, that's what the column, 'R-squared', tells us. This measures the extent to which the regression line is a good fit for all the scattered dots on the chart. The closer that R-squared is to 100, the better the fit. True, with investment data, values rarely get much above 30 (which would tell us that 30 per cent of the dependent variable's returns are driven by the independent variable). But in our study each R-squared value is less than 1 per cent, which is another way of saying that, to all intents and purposes, levels of leverage tell us almost nothing about future stock returns on average.

Yet that's not the whole story. Levels of leverage at one point may say little, but what about companies that consistently carry a similar level of cash or debt? Is that consistency a useful indicator?

This is where we came in and the answer is a tentative 'yes'. It is remarkable how many companies carry net cash year in, year out - 63 of the 79 FTSE 350 companies that had net cash in 2004 also had net cash in 2014. True, we have not measured what happened in the interim. Some may have called on shareholders to subscribe new equity - as in the case of Barratt Developments (BDEV). But, most likely, the companies consistently held net cash for one or more of four reasons:

■ It was in the nature of their business to do so (for example, fund managers, such as Rathbone Brothers (RAT) or Man Group (EMG);

■ They were young-ish companies whose cash profits were too limited to allow them to carry much debt yet they wanted to be able to seize investment opportunities should they arise (pharmaceuticals group Shire (SHP) or microchip designer Imagination Technologies (IMG);

■ Directors owned a lot of the company's shares (engineer Renishaw (RSW) or copper miner Antofagasta (ANTO);

■ They were high-quality companies that consistently generated more cash profits than they could invest or reasonably distribute (aero engine maker Rolls-Royce (RR.) or microchip designer Arm (ARM).

Obviously, companies in the last of those four categories are the most enticing. Stock-pickers may even want to take that train of thought to the next station. They could use the raw data for our findings, which is on an Excel spreadsheet which can be accessed via the link below and identify the high-quality cash generators; not just Rolls-Royce and ARM, but also the likes of Aveva (AV.), Oxford Instruments (OXIG) and Victrex (VCT). The next step is to reckon which ones' shares are still fairly valued and fine-tune a portfolio from those.

However, whether their returns would beat the average of the 45 that had net cash in both 2004 and 2014 is far from certain; perhaps too far for the effort to be worthwhile. Maybe we should just accept that 'cash is king' and leave it at that.