Global Multi-Asset Viewpoint
When the Fed raises the price of money from zero, it changes everything. For the past 10 years, money has essentially been free. In certain countries, they were actually giving it away (and still are). In the U.S., the Treasury borrowed short term at around 0.05% for most of the past decade. Highly-rated corporates and speculators funding levered investments borrowed at a LIBOR rate of 0.25%. Households were never able to borrow as cheaply, but some floating-rate mortgages fell as low as 2.5%, the lowest rate ever.1
What happens when money is free? Everyone (households, businesses, governments, speculators) lines up to borrow some. In this cycle, it has been mostly businesses, governments and speculators, as households were too recently burned by their housing borrowing binge of the early 2000’s to want to re-lever. When money is free, debt rises sharply because interest payments are so low, no one has trouble servicing it (Display 1). In addition, the price of assets rises: when funding is near zero, any asset yielding anything looks tremendously attractive.
Real estate, with rental yields at record lows of 4.0%, looked very attractive; utility stocks, with a dividend yield of 3.7%, also attractive; junk debt, at 6.4%, staggeringly attractive; start-ups with big payoffs in 10 years, never looked better on a present value basis; LBO deals at 11x EV/ EBITDA, what’s not to like? Money is so cheap they’re practically begging you to borrow.2
Unfortunately, this fairy tale environment has come to an end. The end of free money has arrived. We are entering a new regime for the economy and for markets. The consensus appears to believe that there are no excesses in the economy, and therefore that a recession will not happen for at least two years and, importantly, since we cannot have a bear market without a recession, stocks must be owned and the dips bought. We disagree. With the end of free money, we believe:
1) Rising rates will expose the cost of carrying excessive debt—earnings will disappoint and defaults will rise more than expected;
2) The higher cost of debt will slow the economy—even though it appears nothing can stop growth today because of the global synchronized economic boom, the U.S. tax cut and fiscal spending ramp; and
3) Risky assets will re-price lower— not because of a 2020 recession, which is somewhat too far out—but because the rising price of money means all assets are less attractive (all else equal).
Below we expand on the impact of higher debt servicing costs on the corporate sector (via EPS and defaults), the economy, and asset prices.
Rising Debt Servicing Costs: Impact on EPS and Defaults
Our analysis indicates that U.S. corporate debt has shot up 50% from its cycle low nearly 10 years ago and now stands at $9 trillion. This represents a leverage ratio of 5.6x (i.e. debt is 5.6x EBIT), at levels last seen in 2007 just before the Financial Crisis. On the other hand, with record low borrowing costs, servicing record debt has been manageable for U.S. corporates. Interest expense has remained relatively low at 19% of EBIT. This is not as low as at the prior cycle bests of 15-17% in 1995, 2005 and 2015, but is much better than stressed periods such as 2001 and 2009 when interest expense rose to 30-45% of EBIT.3 However, our work shows that every 100 bps of rate hikes will increase interest expense as a share of EBIT by 200 basis points (i.e. from 19% to 21%) such that by the end of 2019, interest expense will rise to 22% of EBIT – the same level reached before the 2008-09 recession.4 (Display 2) The increase in interest expense is driven by the fact that over 30% of corporate debt is linked to short term rates and about a fifth of fixed rate debt will roll over at higher rates each year.5 Ability to pay is one of the key drivers of default rates and interest expense rising to 22% of EBIT points to U.S. high yield default rates rising from the current 3.5% to 4.2%. A 4.2% default rate implies high yield spreads of 490 basis points (up from the current 360 basis points).6
This outcome could be worse given that investment grade debt is of the lowest quality since 1990, with almost half of investment grade borrowers rated BBB, two notches above junk. In addition, by 2020, EBIT is likely to fall while debt will likely still be rising and unless the Fed actually cuts rates quickly enough, interest expense could rise to 27% of EBIT, which could drive the default rate above 6% and spreads above 800 basis points.7
Simultaneously, these rate hikes and interest expense increases would hit margins and earnings, which does not appear to be in consensus estimates. Our work points to a 3% hit to S&P 500 EPS in 2018 and 2019, from higher interest expense. This represents a roughly 35 basis point annual hit to margins in 2018 and 2019.8 And yet, the consensus forecasts margins to increase by 25 basis points in 2019 compared to 2018.9 This alone could cause EPS to grow at only 5-7%, rather than the 10% currently expected by the consensus for 2019. Some disappointment is also possible for 2018 EPS but will likely be swamped by the enormous boost from the tax cuts, strong global growth and the weaker dollar.
Net-net, 2018 and 2019 rate hikes will increase interest expense and corporate defaults and cause corporate spreads to widen and margins and EPS to miss expectations.
Rising Debt Servicing Costs: Impact on Economy
Given the apparent on-going global synchronized growth boom and U.S. fiscal stimulus, it is no wonder that economists and market participants hold a very positive view on U.S. growth in 2018. We share that view. The consensus expects U.S. growth of 2.8% this year and our forecast is 2.9%. However, it is striking that the consensus expects growth to remain exceptionally strong in 2019 and 2020, at 2.4% and 2.1%, respectively. These forecasts mean the economy would grow 0.9% and 0.6% above its potential growth rate of 1.5% in the 10th and 11th year of an economic expansion. In our opinion, this is extremely unlikely, as it implies that the Fed’s hiking cycle of 200 basis points by 2018-end and 275 basis points by 2019-end will have had a near-zero impact on economic activity. Remember it took zero rates over the last 8 years to get the economy to grow at a 2.2% pace.
On the contrary, we expect growth to moderate significantly in 2019 and in 2020, as the Fed’s actions are already beginning to bite and will continue to do so over the next two years.
Conceptually, we believe one of the better ways of describing the degree of policy tightness is to compare the real fed funds rate to the Fed’s estimated neutral rate.10 For nearly 10 years, real policy rates were 100-200 basis points below the neutral rate. Today, because of 150 basis points of tightening from the Fed, policy is now neutral (which is clearly a function of the very low real neutral rate or r* – we use the Laubach-Williams estimate of r* which is currently 0%). Historically, going from very easy to neutral policy has been sufficient to cap growth and drive it back towards trend. And 100-130 basis points of “tightness” (i.e. rates above neutral by 100-130 basis points), which is where we will be by end-2019, has historically been sufficient to drive the economy into recession.
More practically, higher policy rates act to restrain growth through multiple channels and are visibly doing so already:
• Housing investment: The expectations of higher policy rates has caused a doubling in 10 year yields in the past two years, which has driven mortgage rates to the highest levels in nearly four years.11 As a result, housing activity has slowed, with home sales flat for the past year. We expect housing investment to go from contributing roughly 15 basis points to growth over the past two years to detracting just over 5 basis points over the next two years.12
• Consumer spending, though in large part a function of disposable income, is also affected by higher interest rates, which impact auto loans, credit card debt and home equity loans. So far those interest rates have only risen 80-135 basis points from their lows. We would expect another 150 basis points of rate increases over the next two years.13 Based on historical behavior, every 100 basis points of rate hikes likely detracts 30 basis points from consumer spending.
• Corporate activity is also impacted by higher policy rates through a variety of channels. Capital spending and investment tends to be negatively impacted initially, via the reduction in profits that comes from higher interest expense, and then, via the lower sales and profits that come from falling housing investment and consumer spending, described above. This has yet to occur, but higher interest expense will likely increasingly impact corporates in 2019. Importantly, higher policy rates also lead to reduced bank lending to corporates. Historically, corporate lending standards stop easing a couple of quarters after rates begin to normalize, and then begin to tighten 6-8 quarters after policy rates reach neutral. Again, we have not yet seen this but expect that the Senior Loan Officer Survey, when released for the first quarter of 2018, should indicate a reduction in easing of credit standards, which would portend a tightening of standards in 2019 (based on the historical 6-8 quarter lag). So capital spending by corporates, which is always partially funded by internal cash flows but also by debt, should stay strong this year but will begin to reflect less easy credit standards by 2019.
Rising Debt Servicing Costs: Impact on Asset Prices
It is well understood that the Fed did create conditions for a borrowing binge, hoping to stimulate economic activity. Indeed, one of the stated objectives of free money was to encourage increases in asset prices in order to generate a wealth effect, so that households would increase spending. In that sense, free money succeeded beyond expectations. As we have previously shown, all assets benefited from massive inflows as households, businesses and speculators flowed cash and reallocated to all other asset classes. As a result, most asset classes are sitting today at record or near- record valuations (see Display 3). At no other time in history have so many asset classes been simultaneously overvalued: a bubble in everything, as The Economist magazine cover recently proclaimed. When you are getting “robbed” at the bank getting 0% return on your savings, long term bonds yielding 1.5-2% look appealing, as do corporate bonds yielding 2-3%, junk bonds yielding 6%, stocks yielding 2-4%, and so on. As mentioned earlier, when you can borrow at near zero (and expect to for a long time), paying up for businesses, as private equity firms have been doing at a record pace, or seeding venture deals for companies that are years away from positive cash flows, makes sense.
However, the world has changed. Money and Residential Housing1 is no longer free: cash is now yielding nearly 2% (a 6-month Treasury bill yields 1.91% and a 12 month one 2.08%).14 That has already driven 10-year yields back up to nearly 3%, though investment grade bonds still only yield 3.8% and stocks still only yield 2%.15 But as inflation approaches 2%, a bit earlier than the Fed expects, cash rates are likely to keep climbing and will likely approach 3% in 12 months.16 Since a bond has both duration and credit risk, it will need to yield more when cash pays 3%, than when cash pays zero, given that cash has no duration or credit risk.
Similarly, large cap equities, which currently pay 2% dividends, clearly look a lot less attractive when cash is 3% than when it pays 0% – particularly as it is very unlikely that the dividend growth will suddenly accelerate to make up for the lack of a dividend yield premium over cash. (Display 4)
For this reason, we are of the view that a reset in asset price valuations is likely to occur as the Fed continues on its tightening path. And importantly, this valuation reset will not necessarily come in anticipation of a recession in economic activity and earnings, as a recession might still be 18-24 months into the future. It will occur because the opportunity cost of investing in longer duration assets has gone from zero to 3%.
In a way, the risk premium has gone from an infinite multiple of cash returns to barely 1-2x cash returns – in many cases that is not sufficient given the massive risk differential between long duration or illiquid assets and cash. This applies to speculators as well, who could earn carry while speculating, on a leveraged basis, on the capital appreciation of many long duration assets: getting paid to wait – does it get better than that?! Now the carry for many such levered speculations has been eliminated: it has become a pure bet on capital appreciation. There is likely to be less capital willing to take these types of bets until a valuation reset restores carry to a level sufficient to justify the risk. Individual investors have shown this willingness to borrow to invest, by taking more margin debt than any time in history. The negative carry will soon make it less attractive to hold long equity positions and give them less tolerance for short- term downside.
In our view then, the end of free money as the Fed raises rates to 2% and beyond is a regime shift that the markets do not appear to have discounted yet. The tailwinds of easy money are turning into headwinds as the cost of servicing record corporate debt hits corporate profits and forces weaker, highly levered companies into default. In addition, higher borrowing costs are already impacting economic activity in housing and will likely impact consumers and corporates, particularly after the boost from the tax cut and fiscal spending passes. Ultimately, economic growth is likely to slow more and earlier (2019 rather than 2020-21) than the consensus believes. Lastly, even though everyone seems to believe that “if the recession is not around the corner, you gotta own stocks,” the surprise may very well be that asset valuations reset well in advance of a 2020 recession as the price of money climbs back to more normal levels.
The scenario we lay out will likely unfold over the coming quarters. As a result, we believe the defensive stance we have favored year to date continues to be warranted in the U.S. (and China), with better value in eurozone and Japanese equities and local rates in emerging markets. Against this cautious backdrop, we do expect a stabilization both in U.S. growth expectations as the stimulus kicks in for the second quarter, and in trade war fears as the war of words turns to negotiations. This may temporarily stabilize sentiment in equity markets until the theme of The End of Free Money reasserts itself, as inflation and Fed rate expectations both continue to climb.