“The old normal shows signs of life
- Globalisation delivers a fall in price level masquerading as deflation. Both secular and cyclical deflation forces are fading.
- We update our long list of determinants of the low real rate era. Bernanke’s “global savings glut” obviously has a place.
The three big picture inflation supports
Cyclical, secular and survivorship
We can be critical of the different ways output gaps are calculated, the numbers themselves and their usefulness as a single measure for encapsulating spare capacity in economies. Nevertheless, the reduction/elimination of slack that they signal, apart from being encouraging in its own right, should help towards resolving the question over how much of the “lowflation” experienced has been cyclical and how much secular.
Even when it comes to secular, long term trends, these shouldn’t necessarily be misconstrued as meaning a permanent shift to a new inflationary state. Whether it be the deflationary influence of globalization or the internet, to the extent that this means greater competition (so reduced pricing power), then it does perhaps reduce inflationary potential “permanently”.
However, in a shift from closed economies to open economies (globalization) or from weaker price discovery to stronger price discovery (the internet), a large part of the impact on prices is a one-off adjustment in the level not a permanent reduction in the inflation rate. It just looks like the latter because it doesn’t happen all at once. Inflation should firm if the pace of globalization slows.
Chart 2 suggests globalization is at least experiencing a pause. It shows the extraordinary shift in the openness of the global economy since the 60s but a leveling off in the trade share of GDP in recent years. And, as Governor Carney of the BoE has warned, “deglobalization” (an ugly word for an ugly concept) would threaten a meaningful build-up of inflation pressures.
Perhaps the last line of defense for inflation, as measured, is “survivorship bias”. If economies open up to trade with each other, production gravitates to their respective comparative advantages and (in principle) output is boosted and prices fall. In advanced economies, we have become used to falling goods prices. But, as Chart 3 illustrates simply, if goods prices fall and services prices rise steadily over time, then the overall inflation rate will rise because the index weighting for goods will fall, unless the relative price change prompts a consumption shift from services to goods.
Whether it be this “survivorship bias” or the tendency of economies to consume proportionately more services as they advance (and as their populations age), Chart 4 shows the mild but meaningful shift from goods to services in CPI baskets.
We suggest that perceptions of r*, the neutral real policy rate consistent with growth at trend and inflation at target, have been framed by the experience of a prolonged period of economic slack and an even longer period of globalization. The impacts of both on inflation are probably fading and the real policy rate required to keep inflation pressures in check will likely rise gradually to a considerably higher level than currently priced.
Real rate drivers – the usual suspects
It is worth periodically rounding up the “usual suspects” cited as causes of the low real rate world we have been in. Here we list suggestions from a variety of sources and throw in a few of our own. We do not claim that it is exhaustive and readers would no doubt add and subtract from what we have below.
Thinking in terms of potential longer-dated real rates drivers – those shifting the supply and demand for savings and investment – it is perhaps useful to split them into those drivers that might have shifted the savings curve and those that might have shifted the investments curve.
Most items we list are self-explanatory and we do not want to go over well-trodden ground in a lot of detail before getting to our main contentions. However, some of the drivers we identify should actually be broken-down into arrays of sub-drivers. In particular, we suggest that there are many facets to the apparent change in capital/labour preference that has subdued capital investment, so we carve out a sublist for that driver.
Causes of investment curve shift to the left?
- The long shadow of the crisis – reduced expected real returns, greater uncertainty over those expected returns or greater risk aversion to that uncertainty
- A decline in innovation, reducing opportunities
- The cost of equity capital has fallen, but nothing like as much as the risk free rate.
- Falling prices of investment goods (and inelastic demand).
- Capital/labour substitution – replacing the former with the latter.
Causes of savings curve shift to the right?
- The “Global Savings Glut” (GSG), especially imported savings from reserve accumulators.
- Demographics – a falling dependency ratio. Workers can save more because they are supporting fewer dependents.
- Precautionary savings accumulated because of crisis.
- Rising inequality raising the average propensity to save.” – source Bank of America Merrill Lynch
We would like to add a couple of comments to the above relating to the GSG theory put forward by former Fed president Ben Bernanke relating the reasons for the Great Financial Crisis (GFC). Once again we would like to quote our February 2016 conversation “The disappearance of MS München” on this subject:
“The “Savings Glut” view of economists such as Ben Bernanke and Paul Krugman needs to be vigorously rebuked. This incorrect view which was put forward to attempt to explain the Great Financial Crisis (GFC) by the main culprits was challenged by economists at the Bank for International Settlements (BIS), particularly in one paper by Claudio Borio entitled “The financial cycle and macroeconomics: What have we learnt?“:
“The core objection to this view is that it arguably conflates “financing” with “saving” –two notions that coincide only in non-monetary economies. Financing is a gross cash-flow concept, and denotes access to purchasing power in the form of an accepted settlement medium (money), including through borrowing. Saving, as defined in the national accounts, is simply income (output) not consumed. Expenditures require financing, not saving. The expression “wall of saving” is, in fact, misleading: saving is more like a “hole” in aggregate expenditures – the hole that makes room for investment to take place. … In fact, the link between saving and credit is very loose. For instance, we saw earlier that during financial booms the credit-to-GDP gap tends to rise substantially. This means that the net change in the credit stock exceeds income by a considerable margin, and hence saving by an even larger one, as saving is only a small portion of that income.” – source BIS paper, December 2012
Their paper argues that it was unrestrained extensions of credit and the related creation of money that caused the problem which could have been avoided if interest rates had not been set too low for too long through a “wicksellian” approach dear to Charles Gave from Gavekal Research.
Borio claims that the problem was that bank regulators did nothing to control the credit booms in the financial sector, which they could have done. We know how that ended before.” – source Macronomics, February 2016
Indeed, conflating financing and savings is the main issue when it comes to the GSG theory. From a “Wicksellian” perspective, one would argue that low rates for too long leads to mis-allocation of capital. For instance if one looks at CAPEX expenditures in US High Yield since 1997, one can see in the chart below from Bank of America Merrill Lynch that prior to the onset of the GFC, capital raised through bond issuance went into more leverage thanks to a buying spree with Acquisitions/LBOs. Of course a feature of a late credit cycle does lead to seeing more LBOs and acquisitions:
– graphs source Bank of America Merrill Lynch
As we pointed out in our October 2017 long conversation relating to inflation entitled “Who’s Afraid of the Big Bad Wolf?“, we had over-inflation of asset prices and too low inflation thanks to the “Money Illusion”. The Fed, subdued inflation expectations and inflation with its various QE iterations. We indicated at the time:
“Credit cycles die because too much debt has been raised
When it comes to credit and in particular the credit cycle, the growth of private credit matters a lot. If indeed there are signs that the US consumer is getting “maxed out”, then there is a chance the credit cycle will turn in earnest, because of too much debt being raised as well for the US consumer. But for now financial conditions are still fairly loose. For the credit music to stop, a return of the Big Bad Wolf aka inflation would end the rally still going strong towards eleven in true Spinal Tap fashion” – source Macronomics, October 2017
Financial conditions remain very loose and with the fiscal boost coming from the Trump administration, no wonder the Fed is becoming more hawkish. You have been warned.
But returning to real rate drivers, Bank of America Merrill Lynch in their note highlight what has mattered most for the “Money illusion” to take place:
“What has mattered most?
Over the past ten years, bond market participants would almost certainly cite risk-free bond buying by central bank reserve accumulators and the duration extinguished by quantitative easing, mitigating the impact of heavy government bond supply as the crisis lifted debt/GDP levels.
However, real rates were already in long-term decline well before the crisis. Taking a longer time frame, a Bank of England Working Paper by Lukasz Rachel and Thomas D. Smith (No. 571, “Secular drivers of the global real interest rate”, December 2015) claimed to be able to account for 400 of the 450 basis point fall in long term real interest rates over the preceding thirty years.
Exhibit 1, clipped from their paper, suggests that the global savings glut has only had a small walk-on part in the unfolding real rate drama.
In their analysis, the big four drivers were: lower growth, demographics, an increase in the spreads between risk-free real rates and the real rates experienced in the real economy (including, for instance, the real cost of equity finance), and the falling relative price of capital. For this last to be a driver of lower real rates one must assume that demand for capital goods is price-inelastic.
They concluded that: “most of these forces look set to persist and some may even build further. This suggests that the global neutral rate may remain low and perhaps settle at (or slightly below) 1% in the medium to long run.” In their forecasts, they see demographics delivering most of this increase, as the Exhibit shows. Chart 6 shows how this relates to an end to the downtrend in the world dependency ratio, with upswings well underway in advanced economies.
Later, we will discuss the interaction between risk-aversion, driving the “spreads” component in the Exhibit, and the global savings glut, in order to contend that this can be a force for a bigger upward adjustment in real rates in the future.
The replacement of capital with labour has many aspects
As before, we will list what we see as potential causes of this phenomenon, rather than discuss them in any detail. They should be self-explanatory. We would also stress that the ordering should not be regarded as signalling an attempt to rank them in order of importance.
Drivers of the trend shift from capital to labour
- Increasing labour market flexibility
- A global “labour supply glut”, resulting from:
- A falling dependency ratio
- Globalisation
- A post-crisis workforce that needed to re-skill and price itself back into work
- A change in firms’ perceived capital-labour risk/cost efficient frontier since the crisis
- Capital intensive goods production has been driven out of advanced economies (their comparative advantage being in services)
- Production reflects consumption. Advanced economies consume fewer goods and more (labour intensive) services
- As a result of the above, the modern advanced economy business is capital-light
Ben Bernanke memorably used the term “global savings glut” to describe excess savings circling the world in pursuit of a return. Admittedly, the world saving rate was a little higher in 2005 (when he coined the term) than now but the overall increase in the world savings rate over time has not been great, while that for the OECD has seen a gentle decline.
The glut that is generally understood to have exerted downward pressure on nominal and real yields refers to the savings recycled from surplus countries to deficit countries as large current account imbalances emerged.
However, there are reasons to be a little uncomfortable with that seemingly axiomatic received wisdom without further elaboration. To the extent that current account surpluses represent the excess savings of countries, there are equal and opposite savings shortfalls in current account deficit countries (notably the US and UK).
Conventional wisdom used to have it that countries with persistent current account deficits needed to pay higher prospective returns to attract and retain foreign capital. Investors have a natural preference for domestic assets, so need to be paid a premium for accepting foreign market risk. Therefore, without any change in global saving, an increase in imbalances would be expected to depress real yields in surplus economies but raise them in economies with savings shortfalls.
Conventional wisdom upended
If the above framework is accepted, then a mild increase in the global saving rate accompanied by the development of large global imbalances would have exerted a downward “income effect” on real yields but an upward “substitution effect” on real yields in economies on the negative side of the global imbalances identity. The net impact on real yields in the US (with the greatest need for imported savings) would have been ambiguous. What has upended this logic has been the change in the risk preferences of the exporters of savings.
When an economy is “self-sufficient” in savings, domestic savers have diverse risk appetites; they invest across the risk spectrum. And when an economy does have a savings shortfall but is financed by foreign private capital, risk appetite also tends to be diverse (FDI, equity portfolio acquisition, etc). Up until the late 90s, this was the norm.
So our contention is that the rise of the reserve accumulators, in pursuit of risk-free government paper, crowded-out risk appetite. The substitution effect became one of increasing risk-free investment appetite surpluses and risk-taking appetite shortfalls. Therefore the nature and sign of the substitution changed.
Chart 9 shows the IMF’s presentation of these global imbalances.
In Chart 10, we regroup and simplify the picture. By unifying European creditors and debtors (which appear above and below the zero line in the IMF layout) we change the outline of the picture a bit.
However, the main thing highlighted by Chart 10 is the surplus share recorded by China and the oil exporters up until the last few years. It’s a major oversimplification, obviously, but these are perhaps the most conspicuous reserve accumulators pursuing risk-free external assets.
But that era appears to be over, insofar as we accept IMF forecasts for the development of imbalances. The present and near future of imbalances looks simpler than the past – Europe will be financing the US.
The flows will be private capital, not public reserves, so have the potential to restore the old regime where a US savings shortfall delivers higher not lower risk-free real rates. This also suggests that even though the spread between US and Euro real rates has widened significantly, there’s more to come.
Was the equity risk premium a casualty of this risk appetite shift?
The BoE working paper discussed earlier discussed widening “spreads” as an important driver of low real rates. No doubt the crisis was a major contributor to a gapping wider in the equity risk premium and a shifting preference towards government bonds will reflect other things, like the aging of the average saver. However, we would suggest that if global imbalances have extinguished risk appetite in the way we have described, then this also played a big part in the late-90s bond-equity “correlation flip” shown in Chart 11 and the widening gap between bond and equity earnings yields.
In this note we have discussed very big picture influences that are likely evolving very slowly. However, the underlying messages seem clear. A closing of the global output gap appears to be coinciding with a waning in the deflationary influence of globalisation, resulting in firming global inflation, or at least a higher r* to keep inflation in check. This would be aggravated if globalisation is actually in retreat.
That a global savings glut depressed risk-free real rates is universally accepted but perhaps the bigger global real yield depressant from global imbalances was the extinguishing of risk appetite – “bad” savings driving out “good” savings. The global imbalances are still with us but the composition is changing in a way that should restore risk appetite and lift US real yields, both outright and (especially) relative to European.” – source Bank of America Merrill LynchWe disagree on the above a GSG was not the reason risk-free rates were depressed, no offense to Bank of America Merrill Lynch but we would rather side with the wise wizards at the BIS than with the reckless wizards such as Ben Bernanke at the Fed and others.
Before we move on to our final charts regarding the “liquidity illusion”, we would like to quote the wise words of Irving Fisher from his 1928 book:
“We may now summarize our findings
- The problem of what to do about our unstable money is one of prime importance
- It has been all but overlooked because of the Money Illusion
- This Illusion is the more serious because every man finds it harder to free his mind of this Illusion as to the money of his own country than of foreign money.
- This Money Illusion so distorts our view that commodities may seem to be rising or falling when they are substantially stationary, wages may seem to be rising when they are really falling, profits may seem to exist when they are really losses, interest may be believed to be rewarding thrift when no real interest exists, income may seem to be steady when it is unsteady, bond investments may seem to be safe when they are merely a speculation in gold. It makes a unit of weight appear to be a unit of value; it hides a chief cause of the so-called business cycle; it has enabled political financiers to employ unsound finance with burdens heavier but with complaints less than if sound finance had been employed; it has led to unjust blame of “profiteers” and of the “money lenders”; and above all it has held back stabilization by concealing the need of it.
- The present fixity of weight of our dollar is a very poor substitute for a fixity of value or buying power.
- By actual index number measurement our dollar rose nearly four fold and fell back to the starting point again between 1865 and 1920.
- Most of the dollar’s fluctuations were while the dollar was a gold dollar (1879-1922).
- They were largely peace time fluctuations; most of them occurred while America was at peace (1879-1898, 1899-1917, and 1918-1922), and much of them when there were no important wars elsewhere (1879-1914 and 1918-1922).
- These fluctuations through serious shrink into insignificance in comparison with the thousand-fold, million fold, billion-fold, and trillion-fold fluctuations in Europe.
- The cause of a falling or rising dollar is monetary inflation or deflation and that , in practice, it is seldom or never necessary to specify that the inflation or deflation is merely relative since it is also absolute as well.
- To go back to the cause of inflation or deflation, the extreme variability of money is chiefly man-made, due to governmental finance, especially war finance, as well as to banking policies and legislation; but also due in part to discoveries or exhaustion in gold mines, and changes in metallurgical art.
- The tremendous fluctuations of money produce tremendous harm analogous to what would result if our physical yardstick were constantly stretching and shrinking but far greater
- a. because the money yardstick is used so much more generally
- b. because it is so much more used in time contracts, because stretching and shrinking are unseen.
- This harm includes a constant robbery of Peter to pay Paul – amounting to sixty billion dollars in six years in the United States alone – a net loss to all Peters and Pauls taken together, confusion and uncertainty in all financial, commercial and industrial relations, constituting much what is called the business cycle, producing depression, bankruptcy, unemployment, labor discontent, strikes, lockouts, class feeling, perverted legislation, Bolshevism and violence. In short the harm is threefold: social injustice, discontent and inefficiency.” – source Irving Fisher, The Money Illusion.
He also added that credit control must always be an important part of any program for stabilization. This is leading us to our final charts relating to the “liquidity illusion” in credit markets.
- Final charts – Always remember that liquidity is a coward
As a reminder, a liquidity crisis always lead to a financial crisis. That simple, unfortunately. In our February 2016 conversation “The disappearance of MS München” on this subject we quoted Dr Jochen Felsenheimer and Philip Gisdakis from their 2008 book Credit Crises:
“Asset price inflation in general, is not a phenomenon which is limited to one specific market but rather has a global impact. However, there are some specific developments in certain segments of the market, as specific segments are more vulnerable against overshooting than others. Therefore, a strong decline in asset prices effects on all risky asset classes due to the reduction of liquidity.
This is a very important finding, as it explains the mechanism behind a global crisis. Spillover effects are liquidity-driven and liquidity is a global phenomenon. Against the background of the ongoing integration of the financial markets, spillover effects are inescapable, even in the case there is no fundamental link between specific market segments. How can we explain decoupling between asset classes during financial crises? During the subprime turmoil in 2007, equity markets held up pretty well, although credit markets go hit hard.” – source Credit Crises, published in 2008, authored by Dr Jochen Felsenheimer and Philip GisdakisOur final charts come from Bank of America Merrill Lynch’s Credit Market Liquidity report from the 12th of September and highlights the “buy-side” versus the “sell-side” imbalance after the GFC and seems to be on every credit investors mind these days, rightly so:
“The ECB has been tapering its QE programme, and asset purchases will finish by the end of this year. Credit market liquidity is becoming more challenging with market participants seeing fewer bids when they need them. We think that when bond market liquidity becomes more challenging, the CDS market is the vehicle to manage risk. Bond trading frequencies have slowed down over the past years; trading volumes in the CDS market are rising rapidly, both in the index and the options market.
The “buy-side” vs. “sell-side” imbalance is the largest it has ever been. In a world of growing buy-side assets but lower street liquidity, sharp corrections are more common. Dealer inventories of corporate bonds are clearly way down on where they were in ’07, but banks also appear more nimble in managing their mark-to-market risks and overall exposures on their securities portfolios.
The CSPP has dominated the European credit market in recent years. The ECB has bought more than €167bn of euro-denominated corporate debt (and this is still growing, albeit slowly). The CSPP has been pivotal in improving the credit market’s strength and resilience. But we can see a shift in market liquidity for the worst in recent months amid rising markets volatility.
Liquidity has been challenging according to the findings of our analysis, and credit investors seem to think that it will deteriorate as the buyer of last resort withdraws and they will be the only buyers left in the market (chart 3).
With inflows drying up and possibly continuing to do so as the rates cycle between US and Europe pushes money out of the latter, liquidity will likely become more challenging (more here).
The trend of selling in secondary to participate on primary is the new norm as inflows have stopped. If macro deteriorates further and investors need to replenish their cash balances to cover weaker fund flows technicals, the bid for bonds would weaken more, we think. No wonder that the key concern for credit investors is that “market liquidity evaporates”; the August 2018 survey reading was the highest since H2 2015 heading into the February 2016 sell-off and amid HY market weakness (on the back of a flare-up in the Greek debt saga, EM risks and oil prices tanking).
Our liquidity indicator at the most distressed levels
Arguably it is difficult to quantify liquidity. So many metrics (bid/offer, turnover, volumes and trade counts), but none of these have the ability to measure “illiquidity aversion” and to what extent risk-aversion has dominated the market. We think the volatility market is providing unique and eye opening insight on the current state of the “illiquidity scare” for market participants.
In our Hold your breath for a bumpy ride note, we highlighted an interesting and rather unique phenomenon that recently emerged in the European credit index options market. Amid significant volatility during the Italian BTP sell-off, we have seen an increase in hedging demand. As a result implied vols have moved well above the levels justified by the underlying spread market. But not only that, as not only have vols underperformed (moved more than) the underlying market, but implied vol skews were heavily bid too, steepening to the highest levels we have seen historically (chart 4).
