This is Ken Rogoff's "debt supercycle" view, by and large:
Some Fed thoughts: QE4 and all that: For months, the mid-September meeting of the Federal Open Market Committee...:
was being telegraphed as the most likely start date of the ‘normalisation’ process. Or, to use another bit of central banker-ese, the day when short-term interest rates would begin ‘liftoff’ from the current range of zero to 25 basis points. There is still time before any decision is made, but the latest utterances from America’s central bankers — corroborated by interest rate futures and options — suggest that a September rate increase is becoming less likely. Bill Dudley, the boss of the New York Fed, said on Wednesday that the argument for moving in a few weeks was ‘less compelling’ than it was just ‘several weeks ago’.
At first glance, this seems strange. After all, Fed officials have constantly been telling us that the schedule of ‘normalisation’ depends on whether the hard US macro data conformed to policymakers’ forecasts — which they have! Whether it’s the latest GDP numbers, jobs, real wages, retail sales, autos, housing, or tax receipts, the US economy continues to chug along at the pace most people on the FOMC expected it would. You may think that pace is too slow to justify rate increases, but it’s not surprisingly slow to anyone except the permabulls and inflation-phobes. (It’s been almost two years since we wrote a piece wondering whether inflation was about to come back, and look how that turned out.)
Of course, central bankers are supposed to make decisions based on what will happen not what already has happened. Predictions are hard, though, and even the Fed admits that the staff forecasts are basically worthless for anything more than a year out. As if that weren’t bad enough, it’s somewhere between extremely tough and impossible to know just how any given policy change affects the economy at any point in time.
It may be tempting to just give up and get out of the monetary policymaking business in the face of this uncertainty. (That’s what we would do, anyway.) Central bankers prefer to take regular polls of people who are paid a lot of money to be right and stand to lose a lot of money for being wrong. In other words, they supplement staff forecasts by looking at the growth and inflation assumptions implied by the relative prices of financial assets. It’s less popular now, but there was a time when the Fed closely tracked forward breakeven inflation and when the European Central Bank watched forward inflation swaps. Nominal bond yields contain expectations of real growth and inflation, plus some premium to entice you away from holding cash. Credit spreads are based on the value of corporate assets and the sustainability of debt payments. Estimates of corporate earnings growth can be extracted from equity valuations if you can figure out a good way to calculate the appropriate discount rate. There’s a lot that affects commodity prices, but one driver is expected global demand, which ought to roughly track the world’s real growth rate.
So when you have bond yields plunging, corporate spreads widening (even excluding energy and mining), stock prices falling, and commodities (except gold) collapsing, it’s possible there is useful information for central bankers to consider. Professor Krugman even went to so far as to say the changes in asset prices amounted to a tightening of policy. That, essentially, was Dudley’s point. As Tim Duy put it:
Critics will loudly proclaim that the Fed must hike in September if only to prove they are not governed by the equity markets. That call will be heard in the next FOMC meeting as well, but it will be a minority view. A thousand point drop in the Dow will not be ignored by the majority of the FOMC. Dismissing what are obviously fragile financial market conditions would be a hawkish signal the FOMC does not want to send. Hiking rates is not going to send a calming message of confidence. That never works. If the history of financial crises has taught us anything, it is that failure to respond with easier policy only adds to the turmoil.
The irony here is that this is the first time since 2008 when the disconnect between the financial markets and the real economy is working against the owners of risky assets. Until now, the owners of capital had been capturing most of the gains of the recovery whilst the people hardest hit by the downturn benefited the least. (That’s what happens when monetary policy tries to compensate for excessively tight fiscal policy.)
Which brings us to Ray Dalio’s latest missive on LinkedIn. His basic argument is that the rich countries haven’t yet made enough progress in cutting their debt loads, and until they do, monetary policy won’t work the same way it has in the past. That leads him to predict that short-term interesting rates won’t rise very much before the next recession and easing cycle, at which point the Fed will once again find itself engaging in large-scale asset purchases to boost growth.
Some people with reading comprehension problems took this as a call to initiate ‘QE4″, but Dalio’s actual position doesn’t strike us as obviously unreasonable. According to the most recent FOMC Summary of Economic Projections (from June 17), policymakers expect nominal short-term interest rates to top out around 4 per cent. Assume rates actually rise that far. A tightening cycle of around 4 percentage points would be a bit below the post-WWII average, although it would exactly match the average since Volcker’s disinflation. It’s also possible that the path implied by the dots would have short-term interest rates peaking somewhere above 4 per cent in 2018 or 2019 before the next recessions leads to cuts. Dalio is simply saying that the dynamics affecting the transmission of monetary policy to the real economy are different now than in 1985-2006, which leads him to think that the future path of short-term interest rates will be shallower than the dots in the chart above.
This view isn’t necessarily that different from what’s implied by market prices. For example, the current 1-year rate is around 36 basis points and the current 2-year rate is around 67 basis points, so the implied 1-year interest rate one year from now is a little less than 1 per cent — significantly below the midpoint of the Fed’s dots for 2016. Since the 3-year yield is only 98 basis points, that means the implied one-year rate for 2017 is just 1.6 per cent — lower than the lowest dot. Even if you take the dots as gospel, it’s pretty easy to justify the current yield on the 10-year note (a little more than 2 per cent) by imagining a world where 1-year rates top out around 5 per cent by 2019, stay there until 2020, and then plunge back toward zero by 2022.
Another perspective comes from looking at options. Douglas Breeden, a professor of finance at Duke’s Fuqua School of Business, has been studying the probability distribution of 3-month Libor rates implied by the relative prices of interest rate caps and floors. (These are derivatives that pay out if rates are either above or below certain levels. The methodology is explained in this paper.) Even before the recent ruckus, the options markets had consistently been implying that there was a roughly 85 per cent chance that short rates would be 3 per cent or less in 2018, while the relative prices imply a two in three chance that short rates will be 2 per cent or less. That’s significantly different from what’s implied by the Fed’s dots. By 2020, the probability that short rates will be 3 per cent or lower falls slightly to around 70 per cent, and drops to a little more than 60 per cent by 2025.
That shift in the probability distribution can mostly be explained by slightly greater odds attributed to extreme outcomes, rather than significantly greater confidence in the Fed’s ‘longer-run’ estimate. For example, market prices imply there’s a roughly 15 per cent chance that 3-month Libor will be at least 6 per cent (higher than it’s been since 2000) in 2025, while the implied chance of that in 2018 is less than one in twenty.
But what’s really interesting is the sharp increase in the odds that 3-month Libor will be less than 50 basis points by the time we get to 2025. Returning to where we’ve been for the past six years is the single most-expected outcome, with markets giving it a one in four chance. That makes sense once you figure that 1) recessions always happen sooner or later, 2) not having a downturn for a total of 16 years would be basically unprecedented in the American experience, and 3) cutting rates from a peak probably no higher than 4 per cent almost certainly means bumping up against the zero bound relatively quickly.
The disagreement between the Fed and Dalio (and the money-weighted-average opinion of traders) may not be as big as some are making it out to be, but the disagreement is real. It can boiled down to how you interpret the following chart, which shows the total stock of debt owed by the private non-financial sector divided by national income:
In the 1950s and 1960s, Americans gradually reacquired their appetite for borrowing after the Great Depression. But that only occurred after the great reflation of the 1940s had pushed down the private debt to GDP ratio by about 100 percentage points from its 1929 level. (That in turn can be attributed to the potent combination of massive fiscal deficits and interest rate caps imposed by the Fed.) Things more or less leveled off by the 1970s. What’s interesting is when you get to the 1980s. Interest rates plunged after Volcker’s disinflation, which made it easier for people to borrow more against the same income. There was a lot of debate at the time about whether the big increase in private debt, much of it used to fund buyouts, M&A, and real estate investment would lead to a crisis and possibly a repeat of the dynamics of the 1930s. (Sound familiar?) It’s worth reading the papers and discussions from the 1986 Jackson Hole Economic Symposium — titled ‘Debt, Financial Stability, and Public Policy’ — to get a flavour for how those arguments went.
As it turned out, both sides were partly right.
The downturn of the late 1980s and early 1990s was worst in areas that had experienced big housing bubbles earlier on, particularly southern California and New England. It also coincided with an enormous wave of bank failures unlike anything since the Depression. America’s first ‘jobless recovery’ followed, and the discontent led to the most successful showing of a third-party presidential candidate in eighty years. The Fed, continually surprised at the economy’s persistent weakness, which it and other economists eventually attributed to a ‘credit crunch’ caused by weak banks and excess private debt, ended up (grudgingly) cutting short-term interest rates by 7 percentage points.
Yet the economy did recover, the credit crunch ended, and the second half of the 1990s were boom years. Moreover, debt levels never really fell, and by 1998 they were breaching new highs relative to national income. Predictions of either depression or faster inflation (which would have helped reduce debt, in theory) both proved to be wrong. Arguably the secret ingredient was the equity bubble, which provided cheap financing for corporates to go on an investment binge while also encouraging households to save less and consume more.
When that burst, and business investment collapsed along with it, there were essentially three ways for the economy to grow: an improvement in net exports, fiscal stimulus, and higher household spending. Option 1 was effectively off the table because of the mercantilism of some of America’s biggest trade partners. Option 2 was tried, to a degree, but was discouraged even by people who now know better. That left the third option. But household savings rates were already lower than ever, while the average American’s net worth had just gotten thwacked by the collapse in stocks.
What happened next should be familiar. (If not, read this op-ed column by Professor Krugman, the Donald Kohn quote here, look back at that chart, and then read this.)
After 2008, debtors defaulted in droves. That’s what’s actually responsible for the bulk of the reduction in US household debt, rather than repayments. The rest became somewhat more conservative in their saving, borrowing, and spending habits. If not for the boom in government-backed student loans — arguably a form of fiscal stimulus by the back door, albeit a relatively inelegant one that’s likely to become an albatross for many — total household borrowing would have plunged even more and consumption would have recovered even less. Companies borrowed plenty to exploit differences in the cost of equity and bonds, but outside the energy sector, the debt wasn’t financing much additional investment.
Put another way, the big drop in interest rates wasn’t enough to get the private sector to borrow and spend as in the past, although it seems to have stopped borrowers from retrenching even further. It also hasn’t been enough to actually facilitate any deleveraging: the ratio of private nonfinancial debt to national income is the same as it was in 2005, and has been flat since 2012. The post-2000 playbook for dealing with asset busts no longer works.
That’s significant. We are now in a world where interest rates can’t go too much lower and household debt service burdens are at their lowest level in decades, yet growth is still sluggish and the current jobs market is still worse than at any point between 1994 and 2008 (at least according to economists at the New York Fed).
The benign interpretation is that higher interest rates therefore won’t matter much. Businesses have locked in low rates for years, if not decades, so changes in policy won’t be felt until after operating cash flows have grown substantially. Down payment requirements matter a lot more than mortgage rates for prospective home buyers. Most US mortgages last decades and don’t float, so higher rates on new loans won’t affect the spending power of current borrowers.
More generally, the appetite for living beyond one’s means by borrowing is probably lower now, on average, than it was before 2008. People who lived through the Great Depression had permanently different spending and saving habits than those who lived earlier, or came later. No matter how much the Fed may want today’s borrowers to cash out their home equity and spend, it’s not happening. Yet the economy is still chugging along. If the transmission mechanism wasn’t working well when the Fed was stepping on the gas, why would we think the same mechanism will suddenly prove potent when the Fed starts stepping on the brakes?
The pessimistic interpretation, which we believe is consistent with what Dalio wrote and what is implied by market prices, is that most of the rich world just doesn’t grow that much unless households and businesses are boosting their debt and eating into their savings. The so-called ‘Great Moderation’ was only made possible by two massive increases in leverage, facilitated by almost non-stop declines in interest rates, plus an equity bubble thrown in the middle. We can’t do that again unless we get a 1940s-style reflation that wipes away private debt burdens and makes future releveraging possible.
Since we didn’t get a Great Reflation, this line of thinking goes, the economy necessarily reverted to its naturally weak state — even after the Fed turned monetary policy up to 11. Thus we have the yawning gap between actual GDP since 2010 and what people expected it would have been before the crisis.
The implication is that any significant cutback in monetary stimulus will quickly cause the economy to sink from steady but mediocre growth into stagnation and then outright recession. One nugget of supporting evidence: practically every central bank that raised rates since 2010 has subsequently reversed course, often bringing short rates down to new all-time lows.
Put into central banker-ese, one possible consequence of being on the wrong side of the long-term debt cycle is that the ‘equilibrium rate’ that ensures stable inflation and full employment is a lot lower than in the past. That limits how much you need to tighten to bring the economy to a halt and also increases the odds of hitting the zero bound in the future, with all that entails.
The Fed clearly agrees with this up to a point, considering how low the ‘longer-run’ policy rate compares with past history. The question is one of degree.
We have no strong view on who’s right, although if forced to choose, we’d side with market prices over central bankers. We probably won’t get an answer for at least a year, since the core issue isn’t when to begin ‘normalisation’ but what the world looks like once policy has finished ‘normalising’.
Better, though, to avoid the entire argument by having a responsible fiscal policy that lets the private sector deleverage, as in the 1940s. (Just not for the same reasons, please.)