Chasing Their Tails..

A Closer Look at S&P500 Earnings

The US economy and the US stock market have both rebounded strongly since the Global Financial Crisis. Unemployment is down, GDP is up, and corporate profits are up more. This piece asks why? In particular, it seeks to answer the question why have S&P500 earnings grown so much faster than nominal GDP or turnover?

The Scale of the Growth

Between 2007 and 2014, US nominal GDP (that is, not inflation adjusted) grew by 20.3%[1]. The companies that comprise the S&P500 lagged this somewhat; total sales increased 17.6% from $9.2trn to $10.8trn[2]. Net profit, by contrast, surged an astonishing 52% from $645bn to $983bn[3]. S&P500 Earnings per Share increased a similar amount, from $67.48 to $103.04[4].

We see three factors driving this surge in profitability: Firstly, companies are paying less tax than they used to; Secondly, lower interest rates have lowered financial expenses; and Thirdly, share buy-backs have boosted earnings per share. It is our contention that each of these factors will be harder to repeat in the future.

Digging into the S&P

Net profit margins for the S&P500 are only a smidgen off all-time highs (see chart below). Excluding energy stocks – where low oil and gas prices have driven down profitability – net margins are higher than they have ever been before.

S&P500 Net Profit Margin, 1990-2015

Source: Bloomberg

US companies’ record profitability is not simply the result of operational excellence. If we look at operating margins (otherwise known as EBIT, or Earnings Before Interest and Tax) then we see a somewhat different picture. As the chart below shows, S&P500 EBIT margins are only a little above their average levels for the last 25 years.

S&P500 Net Profit Margin, 1990-2015

Source: Bloomberg

What is the difference between operating and net margins? Simply, financing costs (i.e. interest payments) and tax charges. The rise in the net profit margin of US corporates, then, is not driven by companies having fundamentally better operating performance, but by lower interest and tax charges. The chart below demonstrates this: in the 20 years from 1990 to 2009 approximately half of all operating earnings were eaten up by interest charges and tax; it is now less than a third.

Proportion of S&P500 EBITDA Paid in Tax and Interest, 1990-2015

Source: Bloomberg, THS Partners Analysis

Not So Taxing?

To calculate the impact of lower tax rates on S&P500 earnings, we have built a model containing every component of the S&P500 for the period from 2005 to 2015. (See Some Notes on Our Methodology at the end of this piece for more details as to how we made our calculations.) We have then looked at three factors on an individual company basis: net profit, tax payable on the income statement, and tax actually paid from the cash flow statement. The table below shows the result of this analysis:

	2007	2014	% Change
Net Profit	645	983	52%
Income Statement Tax	354	406	15%
Cash Tax Paid	347	347	0%
Income Statement Tax Rate	35%	29%
Cash Tax Rate	35%	25%

All figures US $ billions

Source: Bloomberg, THS Partners Analysis

Between 2007 and 2015, large US companies’ declared tax rate dropped 6%. More astonishingly, their cash tax rate dropped a full 10%; in other words, corporates have taken provisions for taxes (likely on repatriation of profits from abroad), but because the money has not yet been remitted, it sits on the balance sheet as a “deferred tax liability”. This, of course means S&P500 free cash flow and consolidated cash balances are flattered by the accumulation of foreign earnings (on which taxes are yet to be paid).

If S&P500 companies had accounted for tax at 35% in 2014, net profits would have been $80bn lower. Of the $338bn increase in total profits, almost a quarter came from the diminished tax rate. Even this understates the boost, however: because corporates paid cash taxes at a significantly lower rate than were accounted for in the income statement, debt loadings for companies were smaller, and so were interest costs. If we were to add to accumulated cash tax savings to debt, and were to charge this at the prevailing 4%, we would see S&P500 companies had a further $25bn boost in 2014. Tax optimisation strategies by corporates, then, accounted for a third of all the increase in S&P500 profits between 2007 and 2015 and were more important to the bottom line than sales increases.

An in-depth look at all the ways that multination corporates seek to minimise taxes is beyond the scope of this piece. However, four examples demonstrate how pervasive tax optimisation strategies are:

1. According to the Australian Financial Review[5], Chevron Inc., lent money to Australian subsidiary at 9% to minimise taxes payable in that country. The yield on Chevron Inc. three year bonds is below 1.5%, so this arrangement had the effect of moving profits out of Australia.
2. The Financial Times details[6] the measures used by Starbucks to minimise tax at the operating subsidiary level by requiring countries to buy coffee beans from a subsidiary in the Netherlands. (The Dutch subsidiary had entered into an arrangement with the government regarding tax rates.) The consequence of this was to move profits from high tax jurisdictions like the UK and move them somewhere lower.
3. Apple has come in for particular criticism for its tax practices, with the bulk of its European profits accruing in Ireland where – according to Bloomberg[7] – it paid less than 2% tax.
4. Pharmaceutical company Valeant Pharmaceuticals engaged in a practice called tax inversion, according to the Financial Times[8], which moved the bulk of its tax liability away from the US (where the corporation tax rate is 35%) to Canada (where rates can be as low as 20%, depending on the province).

If we look at the top ten companies in the S&P500, we can see that all but one lowered their declared tax rate between 2007 and 2014:

	2007	2014	Change
Apple	30.2	26.1	-4.1
Alphabet	25.9	19.3	-6.6
Microsoft	30.0	20.7	-9.4
ExxonMobil	41.8	34.9	-6.9
General Electric	15.1	10.3	-4.8
Johnson & Johnson	20.4	20.6	0.2
Wells Fargo	30.7	30.4	-0.3
Berkshire Hathaway	32.7	28.2	-4.5
Amazon	27.9	n/a	n/a
JP Morgan	32.6	27.0	-5.7
Average			-4.7

Source: Bloomberg

It is – of course – the duty of companies to maximise value for their shareholders, by minimising their tax payments. Nevertheless, the S&P500 companies that we have tracked seem to have taken this to a new level. The chart below shows how the declared (P&L) tax rates for various indices compare to the local tax rates. As can be seen, European and Asian companies pay corporate taxes in line with their corporate tax rate; American companies pay far less.

Declared Taxes vs Local Tax Rate, 2014

Source: Bloomberg, THS Partners analysis Note: all data is for components of indices: UK is FTSE 100, Germany is Dax, Hong Kong is Hang Seng, France is CAC40, Canada is S&P TSX, Spain is IBEX, the USA is S&P500 and Japan is the Nikkei 225

Furthermore, as the chart below shows, the habit of American companies of accruing profit overseas and not remitting it back to the parent does not seem to have spread. Cash taxes paid in most countries are in-line with declared taxes.

Cash Taxes Paid as % of Income Statement Tax Expense, 2014

Source: Bloomberg, THS Partners Analysis Note (1): Japan’s very low cash tax charge is largely the result of a small number of financial companies not paying taxes due to tax loss carry forwards; UK numbers are skewed by the disposal of Verizon Wireless which resulted in a large cash charge in 2014, which had been provisioned in 2013. Note (2): all data is for components of indices: UK is FTSE 100, Germany is Dax, Hong Kong is Hang Seng, France is CAC40, Australia is ASX, Spain is IBEX, the USA is S&P500 and Japan is the Nikkei 225

We have hardly been alone in noticing the extent to which S&P500 companies have pursued aggressive tax optimisation strategies. It is a point of economic principle that firms incur tax liability at the point at which work is done. It is our contention that a number of multinationals are using structures that move the tax liability to places far from where the economic work takes place.

The political zeitgeist will not allow this to continue indefinitely. Whether in the UK, Australia, continental Europe or the United States itself, there is pressure on large companies to “pay their fair share”. We have already started to see legislative moves in Europe (the so-called “Google Tax”, for example), and we do not doubt there will be others.

For the analyst or fund manager looking at the US stock market, this brings an uncomfortable conclusion: at the very least, the substantial boost to earnings from lower tax rates will not recur. And it is quite possible that it will be reversed over the next four to five years, dragging headline earnings down even as revenues grow.

Taking Interest

The post Global Financial Crisis period has seen record low interest rates across the developed world. In the US, this has meant interest rates that have been hovering only just above zero for the past 6 years[9]. Furthermore, Quantitive Easing depressed government bond yields, lowering – in turn –  the rates on corporate bonds.

One consequence of this is that large, solvent corporates’ borrowing costs began to come down. In the end of 2006, BBB rated US corporates (which constitute almost half of S&P500 constituents), paid 5.91%[10] for 10 year debt. At the end of 2014, that had fallen to 3.83%[11]. This fall is almost exactly mirrored by the by the drop in interest rates paid by (non-financial) S&P500 firms, which came down from almost 6% in 2007 to only a smidgen above 4% in 2014[12]. Perhaps unsurprisingly, the falling cost of debt encouraged firms to issue more of it. The chart below shows this effect: while interest rates paid by firms were broadly stable between 2005 and 2010, total debt barely budged; as rates collapsed post 2010, S&P500 firms issued almost $900bn in new debt.

Total Debt and Average Interest Rate, S&P 500 (ex-Financials), 2005-2014

Source: Bloomberg, THS Partners Analysis

Lowering the financial burden of more than $3.4trillion by around 2% has the effect of increasing pre-tax profits by $70bn, and net profits (assuming a 29% tax rate) by just under $50bn. On this analysis, of the $338bn increase in S&P500 net profits, around 15% came from lower interest costs.

It is worth noting that the decline in interest charges did not benefit all companies equally. Some companies, like life insurers, saw their profitability and solvency negatively affected by low interest rates. Firms that make profits based on earning a return on a “float” (such as Property & Casualty insurers or PayPal) were also negatively affected. Furthermore, a substantial minority of US corporates are net cash, and saw income decline. In aggregate, nevertheless, the headline profits of US corporates clearly benefitted substantially from the move to lower interest rates, and used the opportunity provided to them to increase their debt loadings.

The chart below makes this point: for non-financial S&P500 components, both the Debt-to-Equity and Debt-to-EBITDA ratios have moved up sharply in the last decade.

Total Debt-to-Equity and Debt-to-EBITDA, S&P 500 (ex-Financials), 2005-2014

Source: Bloomberg, THS Partners Analysis

Interest rates – one would expect – will not remain close to zero forever. It is widely expected that the Federal Reserve will raise interest rates, albeit only by a small amount, at its December meeting. And when interest rates do begin their, inevitable, move upwards, it will only have a small impact on most companies’ interest costs in the short to medium term. Nevertheless, it is important to realise that the next five years will not see the same $70bn boost to S&P500 companies’ earnings that we have seen in the last five years. Indeed, as with tax rates, it is more likely that interest expense is likely to rise more quickly than sales going forward, dragging on overall earnings growth rates.

Sharing the Profits?

Rising debt levels in US corporates have not funded capital expenditures. (Historically, periods of borrowing have coincided with increased investment: firms typically borrow to build new capacity.) Instead they have been used to dramatically boost share buybacks. The chart below shows the total level of share buybacks – as measured by the cash flow statement – over the past 10 years.

S&P500 Companies, Total Share Buybacks, 2005-2014

Source: Bloomberg, THS Partners Analysis

There is clearly a very high degree of correlation between share buybacks and total debt. And there are good reasons why firms borrowed to buy back shares. Firstly, with few firms running into capacity constraints, and with concerns about demand growth in many developed economies, there was little reason to invest in new capacity. Secondly, with both interest rates and – in 2011 and 2012 at least – valuations low, there was a clear benefit to EPS from buying back shares.

To calculate how much of a benefit there was to corporates from share buybacks, we have made a number of calculations. Initially, we looked at the value of shares bought back (in US$) and compared that to the aggregate market capitalisation of the S&P500. The chart below shows the results of this analysis, and – simplistically- we see that the (value weighted) share count of the S&P500 should have dropped 23% between the beginning of 2006 and the end of 2014.

Share buybacks as Percentage of Market Cap and Implied S&P500 Share Count, 2007-2014

Source: Bloomberg, THS Partners Analysis

However, this overstates the impact. S&P500 companies, it transpires, did not see the dramatic drop in shares in issue that aggregate buyback numbers imply. In fact, looking at each individual company, and measuring the change in shares in issue, suggests far, far fewer shares got bought back than should have been. The chart below demonstrates this: over the last five years, rather than share counts falling by an average of 3% a year, they instead fell by just 1% per annum.

S&P500 Share Buybacks and Issuances and Effect on Share Count, 2006-2014

Source: Bloomberg, THS Partners Analysis

There are three reasons behind this disparity: Firstly, S&P500 companies, particularly those in the technology sector, have been prodigious issuers of stock to employees. Many companies trumpet large share buybacks, the main effect of which is to mop up options to employees. (The current trend of ignoring stock based compensation when calculating earnings seems perverse when S&P500 companies alone spend almost $150bn annually in neutralising the effect of share options.) Secondly, a number of companies have issued shares to pay for acquisitions. This tends to boost the aggregate figures, while not increasing per share data. Thirdly, there are a large number of S&P500 companies that have needed to issue shares. (This mostly happened in and around the Global Financial Crisis, when over-indebted firms needed to pay down debt.)

It is difficult to disaggregate these individual factors, and there is much that is subjective. Certainly,there is no reason to suppose that S&P500 companies will not continue to use their profits to buyback shares in future and therefore it is not unreasonable to assume an ongoing boost from these.

However, it is not possible for firms to continue leveraging up to buyback shares. At some point the debt-to-EBITDA levels reach a limit, and the quantity of buybacks must diminish. Furthermore, when shares were trading on 12x earnings and interest rates low and falling, then the benefits of borrowing to buyback stock were large. With the S&P500 now trading on 18x earnings[13], and US interest rates rising, the benefit must be much smaller. If we hold leverage – as a multiple of EBITDA – constant for 2016 and 2017, and assume multiples remain constant, then the boost to EPS for the next two years will be 0.5% and 0.4% respectively. This is a substantially smaller boost than the c. 1% a year seen between 2010 and 2014.

Pulling it All Together

When we began the process of calculating the composition of S&P500 earnings growth, we will admit that we came with certain preconceptions. In particular, we thought the major contributor to growth was going to be higher than normal margins, the result of firms concentrating more on profits than growth.

The reality is that the boost from improved operating margins – while meaningful – was far from the dominant factor. The biggest driver of S&P500 earnings growth at around a third of the total, was lower taxes. Just behind that, in terms of impact, was sales growth. Of the remaining 35%, 15% came from lower interest payments, and then about 10% each from margin improvements and debt-funded share buybacks. (This is taking a limited view of the cost of share buybacks: if we charge the full cost of mopping up option exercises then we need to lower the ongoing increase in S&P500 earnings per share significantly more.) The chart below shows the breakdown from our analysis.

Drivers of S&P500 Earnings Per Share Growth, 2007 to 2015

Source: THS Partners Estimates

We are, in general, optimists about the world. Our view is that the US economy is likely to continue its expansion, while the Eurozone recovery gathers pace. Furthermore, while the transition from an investment led to a consumption led economy in China may be bumpy, we think it will generally be good for the world (and for US corporates). This should mean that the organic sales growth rate for the S&P500 should accelerate in the coming years.

However, US corporates – and S&P500 earnings growth in particular – are likely to face significant headwinds. We suspect there is little opportunity for tax rates to fall further, and we would not be surprised to see them rise, as governments seek to clamp down on the off-shoring of tax liabilities. Furthermore, interest expense has likely bottomed as a cost, and may now start to rise with interest rates. Finally, S&P500 companies’ shares are no longer cheap, and the opportunity to lever up balance sheets to buyback shares may now be behind us.

Together, these mean that S&P500 companies may not be able to grow earnings at the rate achieved in the past. Modest rises in tax rates or in interest charges could almost entirely absorb the benefit from sales growth over the next few years.

When we look at the S&P500, we see high valuations relative to history, a turning rate cycle, and pressure on margins. This is an unappealing combination.