oggi sul solito alfavill un altro lettura dal titolo apocalittico
![Big Grin :D :D](data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7)
un' altra testimonianza delle distorsioni gigantesche attuate sul mercato attuale
The largest arbitrage ever documented
Posted by
Izabella Kaminska on Sep 14 10:40. National Bureau of Economic Research research claims to have documented the largest ever arbitrage.
It’s contained in a great little paper
published earlier this month and it isn’t a fancy, schmancy accessible to high frequency traders only type of trade.
In fact, you can forget about the concept of picking up pennies in front of a steamroller, because according to the authors Matthias Fleckenstein, Francis A. Longstaff, Hanno Lustig this arbitrage can run to as much $20 per $100 notional amount.
So is it based on the mispricings of some far flung emerging market debt instruments?
Not quite. According to the authors the biggest arbitrage opportunity ever may in fact be based on very prevalent mispricings between US TIPs and Treasury bonds.
As the authors explain (our emphasis):
In this paper, we study the relative pricing of TIPS and Treasury bonds. A simple no-arbitrage argument places a strong restriction on the relation between the prices of these securities. We show that this no-arbitrage relation is frequently violated in the markets. To our knowledge, this arbitrage, which can exceed $20 per $100 notional amount, represents the largest arbitrage ever documented in the literature. Furthermore, the sheer magnitude of this mispricing in markets as deep and actively traded as the Treasury bond and TIPS markets presents a serious challenge to conventional asset pricing theory.
The authors go so far as to imply that the Treasury may be mad to keep issuing TIPS. If they hedged their exposure by buying back outstanding TIPS, they could in the authors’ opinions save themselves as much as $56bn:
These considerations imply that the Treasury has strong disincentives to issue TIPS. In particular, our results indicate that the Treasury could have saved up to $56 billion by buying back TIPS, entering into inflation swaps, and issuing Treasury bonds with the same maturity instead. Hence, from a public finance perspective, nominal debt unambiguously dominates indexed debt.
Whenever the government issues indexed debt, it gives up a valuable option and leaves money on the table at the same time. On average, the U.S. government has to levy $2.92 more in taxes, in present discounted value, to repay $100 of debt issued if the debt is indexed rather than nominal. This leaves us with the perplexing question: Why does the Treasury issue TIPS?
Generally speaking, then, it seems TIPS are undervalued versus regular bonds.
The only factor that actually helps soothe the mispricing is the inbuilt deflation floor in TIPS — hence as deflation becomes a greater possibility, the arbitrage window does seems to tighten. Unfortunately for the Treasury, though, even this doesn’t help its costs, since the inclusion of a deflation floor actually creates an ever growing liability in that scenario.
As to what prompts the mispricing, the authors say there are many factors that possibly contribute. However, some of the key issues remain liquidity and issuance size (supply) — which is obviously much smaller for the TIPS market than it is for the regular Treasury market.
As they explain:
Using a regression framework, we explore the determinants of TIPS–Treasury mispricing. The results strongly point to supply-related factors as key drivers of the mispricing. In particular, we find that TIPS–Treasury mispricing narrows significantly whenever the Treasury auctions either Treasury bonds or TIPS. In contrast, the mispricing widens significantly when primary dealers have difficulties in obtaining Treasury securities and experience increased levels of repo failures. These results argue that the supply of securities in the market may play a much greater role in determining prices than classical asset pricing models indicate.
Therefore it seems that sizeable Treasury bond issuance can close the size of the TIPS-Treasury arbitrage.
What’s interesting, though, is that on the flip side —
repo failures do accentuate the arbitrage.
This is easily explainable, as the authors note:
The strongest results in Table 4 are for the repo fail variable in the liquidity category. The amount of repo fails is highly significant in all of the eight specifications, often with t-statistics in excess of 4.00. All of the coefficients for the amount of repo fails are positive, indicating that as repo fails become more prevalent in the market, the size of the arbitrage widens. Since repo fails are the result of market participants experiencing difficulty in acquiring specific Treasury collateral, these results provide clear evidence that the supply or liquidity of the securities involved in the arbitrage is directly linked to the size of the arbitrage. As far as we are aware, this is the first time that such a result has been documented in the literature.
The greatest of mispricings of recent times, meanwhile,
seem to have coincided with the run up to Federal Reserve’s announcement that it would engage in quantitative easing in March 2009:
Weirdly, though, the Fed’s bond purchase programme itself seemed to drive the mispricing away — despite eating into the level of supply in the market:
This, of course, does tie with reports that Lehman’s repo trades were mainly conducted using TIPS as collateral.
As the United States Government Accountability Office
noted in September 2009:
Lehman Brothers owned TIPS as part of repo trades or posted TIPS as counterparty collateral. Because of Lehman’s bankruptcy, the court and its counterparty needed to sell these TIPS, which created a flood of TIPS on the market. There appeared to be few buyers and distressed market makers were unwilling to take positions in these TIPS. As a result, the TIPS yields rose sharply.
According to the authors, then — by including TIPS in its asset purchases the Fed actually helped to bring the two markets back to equilibrium:
In spite of the relatively small size of the TIPS purchases, relative to the Treasury bonds, this program appears to have dramatically reduced the mispricing gap between TIPS and Treasury bonds. On March 18, 2009, the average amount of mispricing was $8.69. Figure 6 (above) plots the time series of mispricing after the announcement.
This number declined to $6.49 by April 19, the date of the first outright TIPS purchase. The total amount tendered to the Treasury was $15.56 billion, and the total amount accepted was $1.50 billion. The mispricing then declined to $2.08 on May 26 when a second TIPS purchase took place. The total amount accepted at this second purchase was $1.55 billion.
Lastly, it’s also worth pointing out that the authors observe that arbitrage opportunities in other fixed income markets, especially those versus CDS securities seem to open-up specifically when the TIPS-Treasury mispricing is at its widest.
As they note:
These results have several important implications. First, they demonstrate that the forces that allow arbitrages to exist in the CDS/corporate-bond markets may also be at work in the TIPS–Treasury market. Second, finding that both the CDX and CDS arbitrages are related to the TIPS–Treasury arbitrage argues against the view that arbitrages are due exclusively to slow-moving capital. This follows from the fact that the CDX arbitrage requires relatively little capital to implement, while the opposite is true for the CDS arbitrage. Third, the evidence that lagged CDX arbitrages are predictive for subsequent TIPS–Treasury arbitrages raises the interesting possibility of cross-momentum in different types of arbitrages.
Of course, suggesting the government close out its TIPS position is easier said than done. Amongst other things, it would involve a large transaction in the inflation-swap market to stamp out possible deflation exposure.
Finding someone to take the opposite side of that trade, though, might not be all that easy.
Although presumably the first port of call would be Pimco.
Related links:
TIPS are the best – FT Alphaville
Tipsy in the TIPS market - FT Alphaville
The perils of releasing the repo rate - FT Alphaville
Zero support - FT Alphaville
FT Alphaville The largest arbitrage ever documented