The Toxic Feedback Loop: Emerging <-> Money <-> Developed Markets
Extreme Developed Market (DM) monetary policy (read The Fed) has floated more than just US equity boats in the last few years. Foreign non-bank investors poured $1.1 trillion into Emerging Market (EM) debt between 2010 and 2012 as free money enabled massive carry trades and rehypothecation (with emerging Europe and Latam receiving the most flows and thus most vulnerable). Supply of cheap USD beget demand of EM (yieldy) debt which created a supply pull for EM corporate debt which is now causing major indigestion as the demand has almost instantly dried up due to Bernanke's promise to take the punchbowl away. From massive dislocations in USD- versus Peso-denominated Chilean bonds to spiking money-market rates in EM funds, the impact (and abruptness) of these colossal outflows has already hit ETFs and now there are signs that the carnage is leaking back into money-market funds (and implicitly that EM credit creation will crunch hurting growth) as their reaching for yield as European stress 'abated' brings back memories of breaking-the-buck and Lehman and as Goldman notes below, potentially "poses systemic risk to the financial system."
Thanks to Bernanke's comments, the suck out of global liquidity, which most obviously chased high growth, high beta, high yielding EM bonds and stocks has broken these markets in a much more significant manner than would be expected from just risk aversion. The unwind of endlessly rehypothecated positions and derivatives leaves things like foreign-specilator-held Chilean USD bonds massively dislocated from their domestically-held Peso bonds...
Yields on Chile’s $750 million of bonds due 2022 surged to a record 3.71 percent yesterday, part of a global rout as the Federal Reserve Chairman mapped out a timetable for the end the bank’s unprecedented bond buying this week. With the sell-off, an investor who buys $10 million of the bonds and converts them to pesos using cross-currency swaps would get 0.61 percentage point more in interest than similar-maturity local notes. That would provide an extra $602,000 by the time the bonds mature, according to data compiled by Bloomberg.
The last time valuations of Chile’s dollar-denominated bonds fell so far out of line with local fixed-rate debt was in October 2011 as European leaders debated a bail-out for Greece. The distortions in the nation’s bond market are amplified because only three percent of Chile’s local notes are held by foreigners, shielding them from the global capital flight. Historically, yields on dollar bonds have been 0.19 percentage point lower than local notes after swapping.
“It is becoming more attractive to buy in dollars and swap to pesos,” said Rodrigo Blazquez, a swaps trader at Deutsche Bank AG in Santiago. “The pain has been felt more in derivatives where there are more offshore investors. There shouldn’t be much flow of sales in local bonds but we are seeing flows in the swaps.”
But as JPMorgan notes, there are significant EM vulnerabilities and potentially dire feedback into credit creation and growth implications should this continue:
Emerging market assets underperformed their DM counterparts sharply over the past few weeks. We have been tracking the flows behind this underperformance via ETFs: both EM equity and EM bond ETFs continue to see outflows, a sharp contrast to this week’s buying of US equity and HY ETFs.
EM Equity ETF outflows reached $25bn since February, meaning that 2/3rds of the previous inflows since QE3 started on Sep 13th have been unwound so far. Looking further back, since QE1 started at the beginning of 2009, EM equity ETFs had seen cumulative inflows of $116bn up until last February and so far 21% of this previous inflow has been unwound.
EM Bond ETF outflows reached $1.6bn over the past month, from their mid May peak, meaning that 1/3rd of the previous inflows since QE3 started on Sep 13th has been unwound so far. Looking further back, since QE1 started at the beginning of 2009, EM bond ETFs had seen cumulative inflows of $18bn up until mid May and so far 9% of this previous inflow has been unwound.
But looking at more complete Balance of Payments data is important when one tries to assess the size of previous inflows into EM and the vulnerabilities across asset classes and regions.
Both the IMF and IIF produce portfolio flow figures based on Balance of Payments statistics. Figure 1 shows portfolio inflows by non-domestic investors into EM equities as well as EM debt, i.e. purchases of EM bonds and loans by private non-bank creditors. What is striking in Figure 1 is not only the amount of EM debt bought by non-residents since 2010, but also the divergence between debt vs. equity inflows. While non-domestic non-bank investors poured $1.1tr into EM debt between 2010 and 2012, they only bought $220bn of EM equities during these three years and only $80bn during 2011/2012.
The strong demand for EM debt over the previous three years made it easier for EM corporate to issue debt. Indeed, Figure 2 shows how strong net issuance of EM corporate debt has been since 2010.
But strong issuance becomes a vulnerability when demand slows, as a large imbalance between demand and supply quickly emerges. In contrast, DM corporate debt supply is a lot more subdued making DM corporate debt less vulnerable to demand/supply imbalances. Equivalent net issuance data for DM corporate debt from the BIS shows how weak DM corporate supply is currently compared to 2007, at the peak of the credit boom (Figure 3).
And this is a risk for EM. To the extent that the retrenchment of foreign creditors becomes more persistent, EM credit creation could slow sharply, hurting economic growth. The IIF’s forthcoming quarterly EM credit conditions survey is important to watch in this regard.
Indeed, high frequency data on EM corporate debt supply showed a sharp decline over the past few weeks.
Our colleagues in EM corporate research (Yang-Myung Hong and team) are reporting that the EM corporate debt primary market has been effectively shut since the end of May. Month to date volumes stand at just above $1bn for external debt with several deals having to be pulled.
Creditor retrenchment is also affecting money markets in EM. Our ELMI Plus Index yield, which is a weighted average of money market rates across EM markets jumped to 5.53% this week, the highest level since April 2009. The Chinese liquidity squeeze is partly responsible for this.
What is most important for China and EM economies more generally is how this tightening in money market liquidity and credit conditions more generally will affect the economy by perhaps leading to higher lending rates and tighter credit conditions. It is also problematic that this credit shock is hitting the EM at a time when growth is already weakened.
Weakening growth has been creating indigestion in the EM equity space for some time, since 2011. A similar indigestion is now felt in the EM debt space also, risking a credit crunch and even weaker growth in EM.
Which EM regions are more vulnerable? One way of answering this question is to look at the size of previous debt and equity inflows between 2009-12 vs. GDP across regions. The most vulnerable regions to foreign creditor retrenchment is Emerging Europe, where cumulative debt inflows were 10.6% of GDP between 2009-12, followed by Latin America with 7.2%. EM Asia is the least vulnerable with cumulative debt inflows of 2.9% of GDP.
But perhaps the following four questions from Goldman Sachs to Edward Eagle Brown Professor of Economics and Finance at the University of Chicago Booth School of Business' Anil Kashyap on the implications of this 'reach for yield' (and how it ends) are most critical...
Jan Hatzius: If the hunt for yield ends badly, which financial institutions/structures would you most worry about?
Anil Kashyap: The money market mutual funds would be ground zero. If interest rates are lower than 12 to 15 basis points, it becomes very hard for them to operate profitably so they’re going to go to great lengths to find yield. And pension funds that may have written annuities that were calibrated to a different level of interest rates will be in a similar yield-seeking boat. I also worry about the ephemeral nature of broker-dealer funding. During the last crisis, what killed the broker-dealers was the funding drying up.
But the basic business model hasn’t changed at all. All of the primary dealers that make markets in the United States still get much of their funding from lending out idle securities parked in customer accounts. The problem is that if the customer decides to take his account somewhere else, the lender not only loses the client’s business but also must replace the securities that it lent with its own money. This process of customers pulling accounts meant that major financial institutions lost tens of billions of dollars in funding in the one week after Lehman failed. There is nothing stopping a similar dynamic from occurring today.
On the other hand, if a hedge fund blows up there’s not that much leverage and somebody typically steps in and buys them. I guess everybody knows that hedge funds are too crazy to give them a lot of leverage and so, in the end, when they have to be sold, you avoid the fire sale. So, I’m not so worried about hedge funds but I am more worried about money market funds, pensions and broker-dealers.
Allison Nathan: Do these worries pose systemic risk to the financial system?
Anil Kashyap: Definitely yes for the money market funds. If it turns out that the way these funds are reaching for yield is funding the European periphery and then suddenly something goes terribly wrong in Europe they would need to pull out quickly and counterparties of counterparties would lose their funding, similar to what we saw in 2008. The worst case could be that some fund breaks the buck (the fund’s Net Asset Value falls below $1/share).
These funds are largely unchecked relative to what they were before the crisis. Similarly, if there was a big round of panic triggered by something in Europe or elsewhere we could see the funding vulnerabilities of the brokers-dealers again tested. Pension funds probably would not generate a systemic crisis but potentially bad macroeconomic effects if people start finding out that their pensions are more broke than they realized. I don’t think that’s great but it probably doesn’t create immediate panic.
Jan Hatzius: How quickly can these risks– that seem somewhat far off today – materialize?
Anil Kashyap: Risks building on the balance sheets of the major financial institutions can happen more quickly than is commonly appreciated. Lehman really transformed itself in the last 18 months/2 years of its life to take much more risk than they ever had before. They trampled their own risk controls. They started overriding the limits that were normally set and so when they finally went down they were in a massively worse position than if they had run the business as they had up until 2005 or 2006. Whether something like that is going on is harder to detect. And
I think any of the large financial institutions that are heavily reliant on short-term, overnight repo-style funding could still wake up one day and discover that they’ve lost access to their funding in the market.
Allison Nathan: Would such a scenario be more or less damaging today than in 2008?
Anil Kashyap: That scenario could almost be more damaging now than it would have been five years ago. The Fed’s powers have been trimmed in terms of what they can do legally. There is no TARP (Troubled Asset Relief Program, which allowed the US government to purchase assets and equity from financial institutions to shore up banks in the aftermath of the subprime mortgage crisis). Dodd-Frank (law signed in 2010 that overhauls US financial regulation) isn’t fully implemented so not everybody is up to the new capital standards. So, if we did get a bad shock now, I think you could worry that it could be even worse than last time.
Did we seriously learn nothing? Or is it that our 'leaders' and 'planners' wil constantly tease us with what they know we love most - free stuff... money.