2023-06-19 03:30:45 ET
Summary
- NASDAQ 100 valuations are not nearly as extreme as they were at the 2000 peak, but at 40x free cash flows, they are still extremely expensive.
- Such valuations can only be justified by rapid growth, but real sales growth has been trending lower for years and sits at just 2% y/y.
- The risks posed by this combination of expensive valuations and slowing growth are heightened by the rise in corporate bond yields, which suggests significant declines in the QQQ are likely.
The outlook for the Invesco QQQ Trust ETF ( QQQ ) is arguably the most negative it has ever been outside of the very peak of the dot com bubble in 2000. Valuations on the QQQ, which tracks the NASDAQ 100-Index ( NDX ), are not nearly as extreme as they were at the 2000 peak, but at 40x free cash flows, they are still extreme, particularly considering the ongoing growth slowdown that has seen real revenue growth slow to just 2% over the past year. The risks posed by this combination of expensive valuations and slowing growth are heightened by the rise in corporate bond yields, which suggests significant declines in the QQQ are likely over the coming months and years.
Price/Free Cash Flow Ratio Now 40x As Rising Costs Pressure Margins
The reported PE ratio on the NDX according to Bloomberg is 33.2x, which is in the top 15% of valuations seen over the past 20 years. However, if we look at the earnings before extraordinary items, which shows what companies actually earned, the figure rises to 39.1x. If we look at the free cash flows that NDX companies generated over the past year, the stuff that shareholders can see via dividends and buybacks, this figure rises to 40.0x, the highest level seen over the past 20 years.
Free cash flows are actually down 19% over the past year due to surging operating and capex costs. Operating costs have risen by 10% over the past year, twice as fast as sales, largely due to a 15% rise in operating expenses (operating costs less costs of goods sold), which suggests labor costs have been a key driver. According to Bloomberg, the total number of employees at NDX companies has risen 11% annually over the past decade, even faster than total revenues.
Meanwhile, capital expenditure is up a staggering 21% y/y, with its share of sales rising to a two-decade high. The willingness of investors to accept low income in exchange for future growth has allowed tech companies to invest a staggering 9% of sales and fully 45% of their operating cash flows on capital investment.
Despite the surge in costs, NDX's net profit margin remains at 11.9%, which is still almost 2pp above that of the S&P 500. In fact, as the chart below shows, the NDX profit margin is higher than the S&P 500 profit margin has ever been outside of the 2022 all-time high.
There are significant long-term headwinds facing profit margins for US tech companies. For a start, it will be difficult for tax and interest costs to fall any further. At just 5%, tax and net income expenses are close to record lows as a share of operating income, around half of the figure for the S&P 500 ex-financials, and less than a third of its long-term average. With interest rates having risen sharply and the NDX seeing its net debt position deteriorate significantly in recent years, interest costs are highly likely to rise. Regarding taxes, there is a growing risk that policies will be drawn up in years to come to impose windfall taxes on highly profitable tech companies in an attempt to redistribute savings.
From a macro perspective, the US economy continues to suffer from a dearth of savings. Recent figures show that US gross saving as a share of GDP (all the stuff the US economy created less what residents consumed) fell to a 12-year low of 15.7% in the first quarter of 2022, and as profits are a function of overall savings, the available profit pie is shrinking. Total economy-wide corporate profits were equivalent to 64% of gross savings in Q1, almost double their long-term average. That so much of a country's savings to be in the hands of a small number of companies reflects the winner-takes-all nature of the US economy, which has at least in part been driven and sustained by the decline in economic freedom and low corporate tax levels.
The surge in the fiscal deficit and in money supply has been the key driver of the rise in US corporate profits and the simultaneous decline in the national savings rate seen over recent years, due in large part to the rise in social payments funded by newly created money. As I argued in June 2021 in ' QQQ: Borrowing From The Future' , the decline in national savings has real world implications for future investment. As we are now seeing reflected in the decline in tech sector free cash flows, the scarce real resources available for capital investment are proving very costly. Note that while the narrowing of the fiscal deficit over the past year has had a negative effect on tech sector profits, it would not require any further declines in the deficit for margins to decline as the scarcity of savings will show up in continued rising input costs.
This Is Not Your Father's Growth Market
Two decades ago the NDX was a small and rapidly growing market. The average NDX constituent was smaller than the average mid cap stock today, with just $4bn in annual revenue, less than one third of the average S&P 500 company. The average NDX company's revenue now stands at $38bn, even higher than the $31bn S&P 500 average.
It stands to reason that large companies cannot grow at a faster pace than small companies over a lengthy period of time. In 2003 NDX revenues were 7% of the overall US market, and this figure has since risen to 24% as annual revenue growth has averaged 11.8% versus 4.8% for the overall market. If this growth outperformance were to continue over the next 20 years, then NDX revenues would rise to 88% of the overall market, which is of course highly unlikely.
Much more likely is we see a continued slowdown in growth reflecting the mature nature of the companies that dominate the index and the weak outlook for the real economy. The chart below shows the annual growth rate of real NDX revenues over the past 20 years. The figure currently sits at 2%, while the trend rate is around 6% compared to 11% two decades ago.
The relationship between valuations and market size has been on display over the past two decades in the relative valuations between the NDX and the S&P 500. The following chart shows the inverse correlation between the PE ratio of the NDX relative to the S&P 500 and NDX earnings as a share of the S&P 500. As you can see, and as one would expect, the greater the earnings as a share of the overall market, the higher the NDX's valuation premium. The red dot represents the latest data point, and on this basis the degree of overvaluation relative to the broader market sits at 30%, which is the most expensive it has been over the past 20 years. Keep in mind that the S&P 500 is also trading at historically elevated valuations.
10-Year Returns Highly Likely To Be Negative
So what kind of long-term returns should we expect from a market trading at 40x earnings and exhibiting a trend of slowing growth? In a recent article on the Vanguard Information Technology ETF ( VGT ) (' VGT: Tech Sector Valuation Premium Approaching 2000 Bubble Peak ') I noted that the last time valuations were at current levels in terms of price-to-gross profit and price to sales ratios the tech sector posted 10-year nominal returns of -2% and -6% respectively.
This may sound implausible at current trend growth rates. Assuming valuations and the dividend payout ratio remain unchanged, with a dividend yield of 0.7% we should expect to see real returns of 6.7% annually. While this would be significantly lower than the long-term average real return of 10.8% going back to 1985, it would still be reasonable. It could well be the case that the larger and more stable makeup of the NDX currently compared to its history means investors require lower than average returns.
The risk, however, is that we see a continued slowdown in real revenue growth, which I believe is highly likely. A continuation of the trend would imply real trend growth of around 2-3% over the next decade. In addition, real GDP growth looks set to slow sharply. While real GDP growth has averaged 2.1% over the past decade, more than half of this growth was driven by rising employment, with output per worker growing at less than 1%. With the working-age population growth slowing to a crawl and the unemployment rate near all-time lows, real GDP growth is likely to average no higher than 1% over the long term. I would not be at all surprised to see NDX real revenue growth average just 2% over the long term, which would actually be above the real revenue growth rate seen in the S&P 500 over the past two decades.
With 2% real revenue growth and a dividend yield of 0.7%, investors would receive annual real returns of 2.7% assuming no change in valuations. The difference between 2.7% and 6.7% annual growth in terms of the risks posed to the QQQ is huge. For instance, if investors were to require 6.7% annual returns in a market priced for 2.7% returns, the dividend yield would have to rise by 4 percentage points, requiring an 85% decline in equity prices.
Rising Yields Maybe The Trigger As Was Seen In 2000 And 2007
Valuations do not tend to matter much in the short term. The marginal QQQ investor is driven much more by the short-term trend than by any consideration of the long-term outlook. The fact that the QQQ has surged over the past few months alongside a collapse in free cash flows is testament to this. However, as we saw in 2000 and 2008, rising corporate bond yields can often act as a trigger for sharp declines in valuations.
The chart below shows the NDX overlaid with periods when high yield corporate bond yields were 3% or more above their 3-year lows. Such conditions have been prevalent just 15% of the time but have coincided with every major decline in the NDX. If one had sold the NDX only when this condition was prevalent and held it when it was not, they would have seen more than triple the NDX's returns over this period.
There have been times such as in 2002 when stability in corporate bond yields failed to prevent significant declines, and there have also been times such as over the past six months when the market has moved higher despite the surge in bond yields. However, the past six months should not be taken as a sign that bond yields no longer matter to tech investors, but rather that the risks are rising by the day. Rising yields put downside pressure on profit margins while raising the opportunity cost of holding stocks relative to bonds. Investors were well aware of the risks posed by rising bond yields last year, when rising yields moved hand in hand with the NDX's free cash flow yield, but they seem to have ignored the risks again for now.
As a result, the Bloomberg US high yield bond index yield is now 3.4x higher than the free cash flow yield on the NDX, which is a new 20-year high, almost three standard deviations above its long-term average. The rise in valuations seen over the past decade or so has been staggering. In 2013 the free cash flow yield on the NDX was almost 8%, higher than the yield on high yield bonds. It has been the decline in the yield since then, rather than growth in free cash flows themselves, that has driven the bulk of NDX returns, contributing 11% to the NDX's 18% returns over the past 11 years. As an aside, at this historic valuation low in 2012, Microsoft ( MSFT ) traded at free cash flow yields of 13%. The stock has contributed 15% of the NDX's gains since then, and the decline in MSFT's free cash flow yield alone has contributed 12% of these gains.
Rising Volatility And Declining Breadth Are Additional Warning Signs
As if extreme valuations, slowing revenue growth, and rising bond yields were not strong enough headwinds, we are also seeing a rise in implied volatility and a decline in market breadth, two factors that have also tended to precede market tops. The next chart shows the NDX free cash flow yield alongside implied call volatility. While the majority of the time the NDX free cash flow yield moves in line with implied call volatility, we often see a slight divergence near market peaks as speculators pay up to bet on further market advances, enticed by the strong trend. The recent rise in call volatility despite declining NDX valuations is another warning sign.
Regarding market breadth, the majority of the time the percentage of Nasdaq Composite stocks trading above their 200-dma and hitting new highs relative to new lows rises alongside the market. At market peaks, however, the majority of stocks tend to be already in bear markets, and this appears to be the case today, with just 45% of components trading above their 200-dma.
Summary
The QQQ is trading at valuations that can only be justified by rapid growth, but real sales growth has been trending lower for years and sits at just 2% y/y. Investors face the prospect of weak long term returns and the heightened risk of a sharp decline that brings valuations down and prospective returns to reasonable levels. The rise in corporate bond yields adds an additional risk factor, with the early-2000 bear market and the 2007-2009 bear market both occurring alongside the kind of rise in yields we currently observe.
For further details see:
Extreme Valuations, Slowing Growth, And Rising Yields Make Tech Stocks A Dangerous Bet