Where Is My Trump Instability Premium?

All evidence suggests that President Donald J. Trump is not ready to put down the bludgeon. On Monday, Trump signed an executive order to pull out of the Trans-Pacific Partnership, that the United States had signed but not ratified. He then announced his intention to renegotiate Nafta. And he all but declared a trade war against China. Given the architecture of global supply chains, a trade war with China would in effect be a trade war against all US trade partners; or at least those in the Western Pacific. A major disruption of global supply chains is a significant risk factor for global markets. US firms have come to rely rather heavily on offshore production and are themselves at risk.

Yet, the bull run in US equities shows no signs of letting up. The S&P 500 hit another all-time high today. Even before the election, stocks were clearly overvalued. What is going on here?

For months before the election, markets rose when Clinton’s fortunes improved and fell when Trump’s likelihood of reaching the White House increased. Wolfers and Zitzewitz estimated that a Trump victory would reduce the value of global equities by 10-15 percent and significantly increase expected stock market volatility.


Source: Wolfers and Zitzewitz (2016)

On election night, markets initially reacted in line with the prediction. But then a strange thing happened. Markets reversed course within hours and the great Trump rally began.


Source: New York Times

The Trump trade is being justified by the promised tax-cuts, infrastructure program and pro-business agenda. But these were common knowledge well before the election. Why did markets change their mind?

I had a very interesting conversation about this with the historian Adam Tooze. He said he was not surprised. In his view, financial markets are reflexive in that market participants’ subjective beliefs determine market outcomes which in turn shape participants’ beliefs and so on. In Soros’ formulation,

The participants’ views influence the course of events, and the course of events influences the participants’ views. The influence is continuous and circular; that is what turns it into a feedback loop.

As I understood it, Tooze has a thicker notion of reflexivity in mind. Specifically, market participants, strategists and commentators construct narratives to make sense of market developments. These narratives gain currency though a complex intersubjective process that is only vaguely comprehensible. They dominate the discourse for a while and at some point that cannot be predicted in advance, they relinquish their hold on the collective imagination in favor of another narrative.

This is pretty much as far as it gets from modern asset pricing. The central insight of modern asset pricing theory is that investors are compensated for bearing systematic risk and not idiosyncratic risk (which can be diversified away). An asset pricing theory in the modern sense tells us what constitutes systematic risk. A theory is entirely pinned down by specifying a vector of systematic risk factors called the pricing kernel.

For simplicity, assume that we have a single risk factor in the pricing kernel, m. Expected excess return of an asset will then be the product of the asset’s beta (the covariance of the returns on the asset with m) and the price of risk (determined by market-wide risk aversion). In the standard Capital Asset Pricing Model, for example, the price of risk is assumed to be constant and m equals the return on the market portfolio, so that the expected excess returns on a stock or a portfolio of stocks is proportional to its market beta. In contemporary intermediary asset pricing models on the other hand, the systematic risk factor is shocks to the leverage of US securities broker-dealers (Wall Street).

We should, of course, expect political risk to be priced in. Especially in times of heightened expected system-wide political instability—say due to the risk of a near-term trade war between the world’s two largest economies—expected excess returns on risky assets should be high. That is say, asset prices should be lower than otherwise warranted. Where, then, is my Trump instability premium?

I am near-certain that Tooze is onto something when he posits that participants’ emplotment of market developments reflexively drive market movements. But such narrative-driven fluctuations are bound to reverse sooner or later. When the Trump trade finally reverses, we are bound to see a large risk-off as the pendulum swings the other way and the market reprices to give me back my instability premium.


Bonus: Banks are leveraged bets on the economy. Banks stocks therefore tend to overperform the market in upswings and underperform them in downswings (their beta is greater than 1). But that’s actually only a small part of the story. The big part of the story is that because banks borrow short and lend long, the profitability of their marginal loan depends on the term spread. And the term spread has widened dramatically as part of the Trump reflation trade. And then, of course, you have the reassurance of adult supervision in the White House.

banks tspread


A Whopper From the President

Wages have risen faster in real terms during this business cycle than in any since the 1970s,” according to the president. That doesn’t sound credible to anyone aware of the tepid pace of wage growth. As I’ll show, he is not even close.

We date expansions as beginning in the first quarter after an NBER recession and ending in the last quarter before the next recession. We then calculate real wages as the ratio of total wages and salaries (BEA) to total hours worked (BLS), deflated by the headline inflation index (FRED). Figure 1 shows the overall gains in real wage per hour during the expansions under the four two-term presidents since the 1970s. We see that real wages grew 11% during the Clinton expansion, whereas they have only grown at 4.2% in the Obama expansion. Indeed, Obama’s performance is even slightly worse than Bush’s 4.5%.


Figure 1.

But the president did not say wages have grown the most in his expansion; he said they have grown the fastest. Since the expansions are of different lengths—ranging from 24 quarters under Bush to 40 quarters under Clinton—perhaps the president has a point about the pace of gains in real wages?

Not even close. Figure 2 shows the annualized growth rate of wages per hour in the four expansions. We see that although the gap is narrower by this metric, the Clinton expansion still yielded significantly larger real wage gains than the Obama expansion (1.05% vs. 0.59%). And instead of being statistically tied, Bush pulls away from Obama. His expansion saw an annual increase of 0.73% in real wages per hour. Meanwhile, Reagan lags behind at 0.41%.


Figure 2.

The White House itself publishes annual estimates of real wages that are included in the Economic Report of the President. Unlike the BEA’s numbers which are for all employees, these are for blue-collar workers only (“production and nonsupervisory workers”). And because the numbers are annual, we have to make a choice of which years to include. We date our expansions from the first year after the end of an NBER recession and the last year before the next NBER recession. For example, the Clinton expansion is taken to be 1991-2000 since the first recession ended in the last quarter of 1990 and the next one began in the first quarter of 2001. We then calculate the annualized gain in real wages per hour for blue collar workers from the data provided by the White House. Figure 3 displays the results.


Figure 3.

We see that according to the president’s own numbers, blue-collar workers did nearly twice as well under Clinton than under Obama, even as the working class did twice as well under Obama than under Bush. During the Clinton expansion, blue collar wages per hour grew at the pace of 0.73% per annum, versus 0.4% per annum during the Obama expansion. Meanwhile, blue-collar workers got shafted under Ronald Reagan. Their real wages per hour fell by 2.8% between ’83 and ’89.

The bad news is that this has been the worst expansion for the middle class since Reagan. The good news is that the Clintons will be back in the White House soon.


Correction: An earlier version of Figure 2 displayed quarterly growth rates instead of annualized growth rates for real wages per hour in the four expansions.

Appendix. Quarterly growth in real wages since the 1970s.



An Illustrated Guide to the US Financial Cycle

Claudio Borio of the Bank of International Settlements is one of the most interesting and original economists of the day. A key innovation of his is the concept of the financial cycle. The idea is that the excess elasticity of the financial sector has dramatic consequences for real activity. Specifically, the supply of credit to the real economy is much more elastic than macroeconomic models have hitherto assumed or would be justified by macroeconomic fundamentals. In good times credit is plentiful and even very dicey borrowers can obtain credit quite cheaply. In difficult times even worthy borrowers find it hard to secure credit.

In order to empirically capture this boom-and-bust cycle, Borio and others developed a measure that uses filtering techniques. The idea is to isolate medium frequency movements in key indicators: credit-to-GDP ratio, total credit to the private sector, and property prices. Borio showed that the comovement of these indicators captures national financial cycles for a number of countries.

Technically: Borio uses a bandpass filter to isolate cycles with length ranging from 8 to 30 years in these three variables and averages them to obtain the financial cycle. Figure 1 displays Borio’s financial cycle for the United States.


Figure 1. Source: Claudio Borio

I recomputed Borio’s financial cycle with more recent data. Figure 2 displays the US financial cycle from 1976-2015. We see that the financial cycle has turned since Borio calculated it.


Figure 2.

US housing has always been a leading indicator of economic activity. Housing-finance is the primary channel through which the excess elasticity of the financial sector propagates to real activity. In what follows, we will see that a single metric of housing-finance, namely mortgage credit-to-gdp, captures the comovement of the components of the US financial cycle quite well. Figure 3 displays raw and detrended US mortgage credit-to-GDP. We can see the extraordinary boom in the run-up to the Great Financial Crisis. Figure 4 displays filtered US mortgage credit-to-GDP from 1951-2016 (using the same bandpass filter).


Figure 3.


Figure 4.

The US housing-finance cycle has become increasingly coupled to credit-to-GDP (Figure 5). It has long been coupled to property prices (Figure 6) and has become increasingly synchronized with the raw credit cycle (Figure 7).


Figure 5.


Figure 6.


Figure 7.

Figure 8 displays the comovement of the US financial cycle and the US housing-finance cycle as measured by mortgage credit-to-gdp. We can observe three closed financial cycles that can be identified either by the three peaks or the four troughs. Mortgage credit-to-GDP (the US housing-finance cycle) barely rose in the first. Then there was a discernible but mild boom in mortgage lending during the late-1980s financial boom. But in the financial boom of the 2000s the two were phase-locked; so to speak. Note the increasing amplitude of both the cycles and the rigidity of the comovement in the last cycle. The past twenty years have witnessed a coupling of the two cycles.*


Figure 8. The US financial cycle and the US housing-finance cycle.

What explains the coupling of the financial and housing-finance cycles? One word: Securitization. Basically, the extraordinary amplitude of the financial cycle in the lead up to the Great Financial Crisis was the result of shadow lending. Figure 9, 10, and 11 show the contributions of banks and credit unions, US housing-finance agencies (“Agency MBS”), and shadow banks (“Private-label MBS”) respectively. Shadow lending accounted for 90% of the increase in mortgage credit-to-GDP during the housing-finance boom of 2003-2007.

Shadow banks here refers to finance companies, ABS issuers, and mortgage real-estate investment trusts (M-REITS), which are essentially artificial firms created by Wall Street to warehouse the raw material (mortgages) used to manufacture financial assets. Thus, securitization brought expanding dealer balance sheet capacity to the housing market and thereby amplified the US housing-finance cycle.


Figure 10.


Figure 11.


Figure 12.

An interesting question for future research is whether housing-finance cycles are synchronous with financial cycles more generally. That is, is this an American peculiarity or is it true of other countries as well? Another open important question is how Borio’s financial cycle relates to Rey’s global financial cycle which is defined in terms of the comovement of global asset prices.

*We know from Rognlie’s work that the growing share of capital income in total income is explained almost entirely by capital gains on real-estate. That’s a third closely-related cycle.



A Monetary Explanation of the Commodities Rout

Commodity prices have fallen dramatically over the past two years. Figure 1 shows the IMF’s commodity price indices for fuels, metals and agricultual raw materials.

IndicesFigure 1: IMF Commodity Price Indices for Fuels, Metals, and Agricultural Raw Materials.

A popular explanation for the commodities rout is the slowdown in China, and more generally, the slowdown in the global economy. A second explanation, favored by Liberty Street Economics, is the strength of the dollar. Since commodities are priced in dollars, a stronger dollar requires commodity prices to fall in order for markets to clear. The dollar has strengthened is large part because the Fed has embarked on a tightening cycle even as the European and Japanese central banks are easing. Another implication of US monetary tightening is a slowdown in the creation of international dollar credit, in turn implying weaker demand for primary commodities.[1] Figure 2 shows the trade-weighted US dollar Index (Dollar Strength) and US dollar credit to non-bank non-residents (Global Dollar Liquidity), compiled by the Bank of International Settlements.

DollarIndCrFigure 2: Dollar Strength and Global Dollar Liquidity.

Commodities are also financial assets, so that their prices are affected by the market price of risk. Etula (2013) showed that the risk-bearing capacity of US broker-dealers—Wall Street banks—is an important determinant of commodity returns.[2] This is because commodity derivatives are largely traded over-the-counter (OTC); that is, in markets where dealers serve as market-makers. Greater dealer risk appetite implies lower expected commodity returns while increased dealer risk aversion implies higher expected commodity returns. Figure 3 shows Etula’s measure of broker-dealer risk appetite (Effective Risk Aversion).

ERAFigure 3: Effective Risk Aversion (detrended).

In order to understand the contributions of these different factors, I estimated linear models for quarterly changes in Fuel and Non-Fuel commodity price indices compiled by the IMF over 2000-2015.[3] I found that a parsimonious model with only three variables (Effective Risk Aversion, Dollar Strength and Global Dollar Liquidity) explains 28 percent of the variation in Fuels and 51 percent of the variation in Non-Fuels.[4] I also tried other reasonable predictors for which quarterly data is available outside paywalls. None had significant explanatory power. In particular, OECD, Chinese and EM growth rates were insignificant for both indices even at the 10 percent level.

FuelFigure 4: Contributions to YoY% changes in the Fuel Price Index.

Figure 4 shows the decomposition of year-on-year percentage changes in the Fuel Price Index. We see that Dollar Strength and Global Dollar Liquidity have been major factors pulling down energy prices. Still, there is a big residual that presumably contains the large-scale effects of geopolitical and oversupply factors. This is certainly the case with energy prices. In 2012-2015, US shale gas production increased by 10 billion cubic feet per day. In the same period, US oil production rose by 3 million barrels a day. In addition, the Saudis essentially declared a price war on their Russian and Iranian rivals as well as on American oil firms. I have no supply-side predictors in the model, meaning that if commodity prices fell due to negative supply shocks then that variation would not be captured by the model. Indeed, it would very sketchy if it did!

NonFuelFigure 5: Contributions to YoY% changes in the Non-Fuel Price Index.

Figure 5 shows the decomposition of year-on-year percentage changes in the Non-Fuel Commodities Price Index. The model performs much better for non-fuel commodities, where it is able to explain half the price variation in 2000-2015. The residuals here are much smaller, showing that a supply glut has been less of a factor for non-fuel commodities than for energy. Dollar Strength and Global Dollar Liquidity each explain roughly a quarter of the price decline in non-fuel commodities over the past two years. On the other hand, the impact of Effective Risk Aversion has been largely positive over the past two years.

The raw correlation between the fuel and non-fuel price indices in the period under consideration is 71 percent, while that of the fitted values is 98 percent.[5] But the correlation between the two residuals is still 57 percent, meaning that while some of the comovement of two series is accounted for by the monetary factors in our model, much of it still begs explanation. The obvious explanation that comes to mind is market expectations of future demand growth.[6] Since commodity prices are forward-looking, market expectations of future demand growth for commodities is likely the dominant factor driving their residual covariation. And that takes us back to China.


[1] Non-US banks, especially European banks, create international dollar credit by lending dollars to non-US residents.

[2] Etula, Erkko. “Broker-dealer risk appetite and commodity returns.” Journal of Financial Econometrics 11.3 (2013): 486-521.

[3] I also modeled Metals and Agricultural Raw Material Price Indices with similar results. These results are omitted here for the sake of brevity.

[4] More precisely, the independent variables in the model were one quarter lagged, detrended Effective Risk Aversion, changes in the trade-weighted dollar index, and changes in the natural log of US dollar credit to non-bank non-US residents. All independent variables were normalized to have zero mean and unit variance.

[5] This is Hamilton’s method relayed by Ben Bernanke.

[6] I considered the possibility of a supply-side cycle. If there were a detectable supply-side cycle, then the residuals would oscillate about zero. But subjecting the residuals to the runs test failed to reject the null hypothesis of randomly generated errors. This was also the case with Metals and Agricultural Raw Material Price Indices.


The Great January Market Sell-Off Is Not Irrational


Many prominent economists seem to be scratching their heads over the recent bout of market turbulence. The consensus is captured crisply by Capital Economics: “The plunge in global stock markets does not seem to be justified by economic developments.” On the other hand, market participants and other non-academic observers are wondering whether this correction will turn into a bear market before the month is out; if it hasn’t already.

The Policy Tensor is usually more sympathetic to rigorous economists than seat-of-the-pants market observers and traders. But in this case, the economists are demonstrating their professional blinders. It is entirely possible to for a market correction to be justified without a fundamental slowdown in the domestic macroeconomy. And this is indeed what is going on.

The US stock market is not a claim on US GDP; it is a claim on US corporate profits. Profit growth can slow without a slowdown in output growth. The profits of US firms have been falling largely due to the strength of the US dollar. Continued policy divergence among the hard currency issuing central banks imply further dollar strength, thus lowering expectations of US profits in the medium term.

Moreover, while exports are only 13.5% of US GDP, the rest of the world contributes half the profits of US firms. Similarly, 97% of employment growth is now taking place in the nontradable sector, whereas the tradable sector accounts for half the growth in gross value added. So, one can have a robust and tightening labor market while profits margins get squeezed due to global disinflation and a rising dollar.

Indeed, as I have argued before, the US economy is much less resistant to an imported disinflation than an imported recession. In particular, this means that the profits of US firms are much less sheltered against global disinflation than the US economy as a whole is to recessionary headwinds from abroad. Therefore, the impact of adverse global developments on the US stock market is bound to be much more serious than on the domestic economy.

Finally, there are purely market developments that have implications for asset prices independent of the real economy. In particular, innovations in systematic volatility imply asset price adjustments regardless of what happens to GDP and inflation.

So what does the market know in January that it did not in December?

First, China: Policymakers in Beijing are clearly floundering and their ability to stem the panic is increasingly coming into question. The renminbi is depreciating faster than anticipated. Capital outflows from China are accelerating to such an extent that the $3.3 trillion cushion — down from $4 trillion — no longer seems invincible. An astonishing trillion dollars of capital has already left the mainland. Meanwhile, the industrial sector is doing even worse than expected.

Second, instead of stabilizing, the commodities rout has exacerbated. This has worsened the outlook for commodity exporters. Moreover, the continued price declines indicate a greater slowdown in global trade and output than previously recognized. To put it bluntly, the world is doing much worse than was realized in the benign aftermath of the Fed’s liftoff; a decision that looks increasingly ill-judged.

Third, it is now being realized that the banks are much more exposed to the oil rout than was assumed to be the case. All the big banks have reported substantial losses on their loan books to the sector. It is clear that these losses will mount since the price of crude is likely to be depressed in the medium term. But it is not clear by how much and that is obviously very worrying given the central role played by financial intermediaries in powering economic growth.

Fourth, market participants can be more confident that market turbulence is likely to persist. The end of the repression of systematic volatility by the Fed and the reduced liquidity due to tighter regulation of broker-dealers increases the likelihood of market turbulence. This means that risk premia will rise since investors must be compensated for greater volatility, in turn implying lower asset prices. In other words: Even if asset prices were aligned with fundamentals before and assuming that said fundamentals haven’t changed, asset prices would still need to fall to reflect the repricing of risk.

So, yes, even though expectations for US GDP and inflation have barely changed since the December meeting, the market correction is warranted.


Hawks Take Centerstage

ACM term risk premia since 2000

Abstract: The market seems convinced that the Federal Reserve will lift-off in December, 2015. We argue that this expectation needs to be tempered because the economy begs to differ with the hawks. Fatally for the hawks’ case, the Phillips Curve is broken. And since the neutral rate is now exceptionally low and on a downward trend, the Fed’s model risk has increased considerably. The labor market continues to show slack on many indicators including decidedly tepid wage inflation. Moreover, the US economy is not nearly as resistant to an imported deflation as it is to recessionary headwinds from abroad. The baseline scenario continues to be lowflation and stagnation for some time to come. The FOMC is therefore likely to hold fire. And if it does hike in December, it would be running the risk of deflation. A premature exit would harm the recovery that is still underway in the real economy. At the very least, the Fed would be sure to miss its inflation target over the medium term.

Read the research note here (pdf). 


How the US Market-Based Credit System Works

In a recent primary debate, Hillary Clinton suggested that risks in the financial system are now concentrated in ‘shadow banks’. I do not like this terminology. The so-called ‘shadow banking system’ is the very core of the US financial system. It is where money is manufactured and the price and quantity of credit determined. We shall use more neutral terms such as ‘the market-based credit system’ or when the intent is clear, simply as ‘the modern system’ instead. In this essay, my goal is to demystify the system for the reader. In what follows, we shall describe the core institutions and key working parts of the modern system. Readers interested in understanding the system in greater detail should begin with the works of Pozsar, Mehrling and Adrian (see references at the bottom); as well as the excellent treatment by the Federal Reserve Bank of New York.

How money is manufactured in the modern system

In a prison, money is cigarettes; for drug dealers and suchlike, money is cash; for most of us, money is bank deposits. But for those with really big balance sheets—corporate treasurers, money-market mutual funds, institutional investors, central banks and big investment firms—it is paper. What is paper? Paper, or more formally a deposit-equivalent, is whatever can be exchanged at par on demand with near certainty. That is, it is an ultra-safe, short-term asset that can be exchanged without loss for cash or bank deposits on demand. A market-based credit system revolves around the manufacture of paper.

How is paper manufactured? Take a risky asset—a claim on any stream of cash payments such as a mortgage or a standard semiannual coupon bond (which pays a fixed coupon rate every six months)—and break down its price into the price of risk and the price of paper according to the first fundamental equation,

asset = paper + risk.

If you can find someone to bear the risk (and earn the reward) of this asset, you can decompose the cash-flow of the asset into junior tranches of asset-backed securities (ABS) that bear most of the risk and senior tranches that bear little. The senior-most tranche after all the slicing and dicing is paper; stamped AAA for good measure. Let’s work with a concrete example to illustrate the manufacturing process of paper.

Let’s take a pool of 10,000 mortgages. Suppose we create just two tranches of mortgage-backed securities (MBS). The senior tranche gets paid in full unless at least 9,000 mortgages default; the junior tranche gets the rest. Now, the probability that 9,000 of 10,000 mortgages default in any given year mathematically depends on the correlation of the rates of default on the 10,000 mortgages. For uncorrelated mortgages, the probability is vanishingly small. And the higher the correlation, the higher the probability that the holders of the senior tranche will lose money. In the extreme, if defaults are perfectly correlated, then the probability that the holders of the senior tranche get burned is the same as the probability of a single default! In practice, the correlation is essentially zero for prime loans. And as we found out in the course of 2007 and 2008, not quite zero for subprime. Anyway, let’s stay with a pool of prime loans—those that are expected to be repaid with very high probability and have low correlation. Since the probability that the holder of the senior tranche will bear any loss whatsoever is basically zero, the senior tranche is paper.

Overnight paper

There are four kinds of paper in a market-based credit system.

(1) Purely public money, where the asset is public and the state provides a backstop. In the United States, these are Treasury bills (Tbills) issued by the government as well as overnight reserves at the Fed. Other government bonds do not qualify as paper because they are subject to considerable interest rate risk.

(2) ‘Public-private’ money, where the asset is public but there is no public backstop. Secured borrowing (called repo) when the collateral on the loan is Tbills and constant net asset values accounts of government-only money-market mutual funds both qualify as this type of paper.

(3) ‘Private-public’ money, where the asset is private but enjoys a public backstop. Bank deposits insured by the FDIC qualify for this type, as do term deposits in wholesale banks.

(4) Purely private money, where the asset is private and enjoys no state backstops. Repo with private asset-backed securities and CNAVs of MMMFs as collateral qualify for purely private paper.

The following table reports the total amount of paper outstanding in the financial system estimated by Pozsar (2014). Note that the largest pile is purely private paper accounting for a full third of the total; whereas purely public paper accounts for about a quarter; about the same as shadow money created on the backs of public assets. Bank deposits, the flag bearers of traditional banking money, account for less than a sixth.

           Type                Quantity          Share

(1) Purely public          $2.6 trillion     27%

(2) Public private         $2.3 trillion     24%

(3) Private public         $1.4 trillion     15%

(4) Purely private         $3.2 trillion     34%

       Total                      $9.5 trillion  100%

Instead of the traditional ‘originate-to-hold’ banking model wherein the loan sits permanently on the balance sheet of the lending bank, the modern banking system is characterized by the ‘originate-to-distribute’ model that parcels out the risks to those who are most willing to bear it; in the process, manufacturing copious amounts of private money. The manufacture of paper involves warehousing the assets off balance-sheet in artificial firms called conduits in offshore locations; and selling some of their risk to monoline insurers. But the paper itself and much of the risk exposure makes its way to the wholesale market; as is clear from a map of the entire system

The Wholesale Funding Market

At the core of the market-based credit system sits the wholesale funding market. It is here that almost all the paper manufactured in the system circulates. At the center of the whole system are the dealers or market-makers, who stand ready to buy and sell, as well as borrow and lend, both paper and assets in bulk at posted bid-ask prices. Dealers supply market liquidity by quoting a two-sided market and absorbing the resulting order flow on their own balance sheets (Harris, 2003). Indeed, a dealer that insisted on a matched book at every instance would not be supplying liquidity at all. Both the volume of order flow and the dealers’ willingness to hold inventories is thus necessary for market liquidity. As Mehrling et al. (2014) note, “If customers are able to buy or sell quickly, in volume, and without moving the price, it is because some dealer is willing to take the other side of that trade without taking the time to look for an ultimate offsetting customer trade.”

Dealers interface with two types of actors in the wholesale funding market—yield seekers who want the returns that come with risk and cash pools who want paper.

Yield seekers are asset managers whose motto is “beat the benchmark”. In order to earn returns, asset managers seek to bear risk. There are two kinds of asset managers. Real money investors who have their own risk bearing capacity such as pension plans, endowments and other institutional investors. And leveraged investors such as hedge funds and other alternative asset managers. Instead of holding equities that would yield 12 per cent with a good deal of uncertainty, a leveraged beta investor holds a combination of paper and derivatives to make a leveraged bet whose expected return is 12 per cent (the leverage comes from the derivative, not from outright borrowing). Such strategies allow for ‘equity like returns for bond like volatility’. This can be seen from expanding the first fundamental equation (asset = paper + risk) into the second fundamental equation,

asset = paper + CDS + IRS + FXS,

where the derivatives—credit default swap (CDS), interest rate swap (IRS), and exchange rate swap (FXS)—are the prices of credit, interest rate, and exchange rate risks, respectively. The proliferation of leveraged beta strategies has therefore generated brisk business for dealers who are market-makers in the asset, paper and derivative markets.

Cash pools’ motto is “do not lose”. These are usually money-market mutual funds, corporate treasuries, foreign central banks, and institutional investors—those holding such a large pile of cash that it is risky to park it at a bank! Since the FDIC only insures the first few hundred thousand dollars in any individual account, cash pools prefer to hold Tbills or ‘quasi-Tbills’ instead. That is, cash pools prefer even purely private paper to bulk bank deposits because the former is more secure. They also want paper for yield since the fractions of a cent on the dollar that paper yields adds up quickly when you have billions of dollars to park. As corporate coffers have filled up with cash and the assets of money market funds has skyrocketed, the demand for paper has exploded. Pozsar (2014) estimates that between 1997 and 2013, while US GDP doubled, cash pools trebled (from $2 trillion to $6 trillion).

The wholesale funding market consists of the wholesale money market and the wholesale capital market. The wholesale money market is the market for paper where money dealers borrow and lend paper in bulk. In the wholesale capital markets, the market-makers can be thought of as risk dealers or derivative dealers.

The repo market for collateralized borrowing, the interdealer overnight lending market, and the off-shore eurodollar market are all part of the wholesale money market. The main one is the repo market. A repo transaction is best thought of as one party borrowing money from another by posting paper as collateral. A reverse-repo is just a repo with the positions of the parties interchanged. Interestingly, while there were runs on the bilateral repo market as a whole in 2007-2008, there were no generalized runs on the tri-party repo market (where the runs were restricted to specific asset classes). This suggests that the panic was about specific counterparties and specific assets. In other words, the plumbing of the system is more resilient that it looks.

Dealers respond to inventory buildups by adjusting their bid-ask prices. But even matched-book dealers bear liquidity risk because they must repay their creditors even if their own debtors do not repay them. The size of dealer books determines the liquidity in the entire wholesale funding market. In turn, the size of dealer books depend on systematic volatility. Dealer balance sheet optimization is inherently procyclical: Dealers respond to low systematic volatility by expanding their books, which in turn further suppresses systematic volatility.

Dealer behavior is also procyclical with respect to order flow. Specifically, dealers move prices to bring buy and sell order flows more into line with each other, but in doing so they move prices away from their matched-book values. This generates a credit cycle. In expansion mode, the boom is exacerbated. And in contraction mode, liquidity dries up.

Pozsar’s central thesis is that the emergence of the market-based credit system can be explained by the growing demand for paper from cash pools on the one hand, and the increasingly frantic search for yield by asset managers on the other. The former is the consequence of a shortage of Tbills—there are only about $800 billion worth of Tbills in circulation. The latter is the consequence of increasingly difficult-to-satisfy obligations of institutional investors. Basically, pension funds and other institutional investors made promises based on overly optimistic assumptions about returns. They were implicitly assuming that productivity would grow at a faster pace than it has. They are thus desperate for excess returns.

Wholesale funding market

These sources of powerful flows on both ends of the wholesale funding market had two consequences. First, they made market-making in the wholesale market extremely profitable. Second, they lowered the price of credit system-wide. Pozsar posits that this mechanism provides a competing explanation (to Bernanke’s ‘global savings glut’ and Shin’s ‘global banking glut’) of Greenspan’s conundrum (the failure of long rates to rise despite rate hikes). [These competing explanations can be tested in a vector autoregression. Yet another project for the Policy Tensor!]

The Fed’s Reverse-Repo Facility

During and after the financial crisis, the Fed moved in to stabilize the wholesale funding market by buying vast quantities of troubled assets; mostly MBS. In addition it expanded the supply of Tbills. The Fed also created an entirely new form of purely public money. This is the Fed’s reverse-repo facility that provides publicly guaranteed paper to the Fed’s counterparties. Moreover, the Fed has given access to the facility not just to primary dealers (who otherwise enjoy exclusive access to the Fed), but also money-market mutual funds. This serves to ameliorate the problem of the shortage of Tbills. Pozsar expects the facility to expand in a big way and drive out much of the private paper in circulation today. So far, the Fed has said that it is a temporary facility that only exists to give the Fed another lever to control the short rate. The facility is expiring in January 2016. It will be interesting to see which way the Fed decides to go in the December meeting. I wouldn’t bet against Pozsar.

A Dealer of Last Resort

The Achilles heel of the modern system is liquidity risk. This is because while credit, exchange rate and interest rate risks end up on the books of those willing and able to bear them (insurers, real money investors and leveraged investors), liquidity risk cannot be so efficiently disposed. This is the familiar problem of systemic risk arising from multiple equilibria—when there is a run on the system, liquidity evaporates. What matters is your ability to make that margin call, not whether you are ‘fundamentally solvent’. What is required, as Bagehot argued in the nineteenth century, is a dealer of last resort. Someone who is willing and able to step in when others aren’t—a role that J.P. Morgan famously played in the panic of 1907. Bagehot’s advice for the Bank of London was to lend freely on (normally accepted) collateral and charge a high rate of interest.

Pozsar, Mehrling, and Adrian counsel the Fed to play this role. On the other side of the pond, the Bank of England has already made moves in this direction. So far, the Fed has declined to accept private paper. My position is that the Fed does not have much wiggle room. It must either produce enough public paper (say through the reverse-repo facility) to drive private paper largely out of circulation, or it must become a dealer of last resort. Otherwise, the market-based credit system will remain prone to runs.

Selected References

Mehrling, The New Lombard Street: How the Fed Became the Dealer of Last Resort, 2010.
Pozsar, “Institutional Cash Pools and the Triffin Dilemma of the US Banking System,” 2011.
Mehrling et al. “Bagehot was a Shadow Banker,” 2013.
Adrian et al. “Repo and Securities Lending,” 2013.
Pozsar, “Shadow Banking: The Money View,” 2014.
Pozsar, “A Macro View of Shadow Banking,” 2015.