Macro Strategy LIBOR OIS

Transcription

1 Macro Strategy LIBOR OIS What Everyone Is Talking About the LIBOR/OIS Spread The spread between LIBOR and the Overnight Indexed Swap Rate (OIS) has become a big topic of discussion. The LIBOR/OIS Spread ( LOIS ) has risen rapidly, to levels that have often occurred during times of overall market stress. The LIBOR/OIS Spread ( LOIS ) since 2001 In 2007, 2008 and late 2011 it was clear that the spread increased due to true stress in the financial system. That is why so many people are fixated on this spread right now and left wondering if it is a harbinger of doom, or something far less nefarious (as you will see, I think it is simply a response to a relatively unique and rapid shift to less supply of money available specifically for short term LIBOR borrowing with a rapid increase in demand for all short term funding, including LIBOR). The bump in 2016 was related to money market reform which reduced the ability of many funds to buy instruments that were part of the LIBOR calculations. During that time, ICE LIBOR was undergoing other changes as the ICE Benchmark Administration ( IBA ) was working to make the calculation more robust (this was a response to problems in LIBOR that had been exposed and resulted in lawsuits and penalties). This bump resolved itself without any broader impact and was an opportunity for those who identified what was occurring and positioned themselves appropriately. The big question, that we will attempt to answer is should we be worried? Is the rise in this spread a precursor to more bad news for markets? Or is this just an opportunity, like the one presented when money market reform took effect? March 24,

2 That is separate from the question of are rising short dated yields an issue? While the market is fixated on this spread, I am more concerned about rising short term yields (and the impact on the consumer, but that is a story for another day). For now, we will focus on analyzing LIBOR s underperformance and explaining, in the end, why it isn t a big concern for me. What is OIS? Before going further, let s just pause for a second and discuss what OIS actually is? The Bloomberg Description is Overnight Indexed Swap (OIS) is a fixed/float interest rate swap where the floating leg is computed using a published overnight index rate. The index rate is typically the rate for overnight unsecured lending between banks, for example the Federal funds rate for US dollars. I ll be honest here, I have trade many strange and bizarre instruments ranging from simple LIBOR products, to bonds, to credit default swaps, to CLO s, to synthetic CLO s, to EETC s, JLL s (Japanese Leveraged Leases), CPDO s (constant proportion debt obligations), etc., and I have never once traded OIS. My gut feel is that many of the people currently talking about this spread barely know what OIS is, or how or why it is traded. I think I understand enough about why it is a good benchmark, to be dangerous, but am not an expert. So you can stop reading here, or bear with me, as I try and take us through what is going on with my knowledge which is sufficient for these purposes. What Happened to the TED Spread Back in the dark ages of finance (like 2 years ago) people still talked about the TED Spread which is just the difference between LIBOR and T-Bills. It has some flaws as a measurement relative to OIS, but for many of us, it is far easier to understand. So, let s look at the Ted Spread. 3 Month TED Spread versus 3 Month LOIS since 2001 The blue line is the LIBOR/OIS spread and the grey line is the LIBOR/T-Bill spread. Fortunately for me, the charts look very similar. March 24,

3 If, they had been very different, I would have stopped writing right here and gone outside and done some yard work that I m trying to avoid, because it would have been an indication that a deep understanding of OIS is necessary. Since the LOIS and TED spreads are behaving similarly, and have behaved similarly, and I understand T-bills very well and OIS just enough, I am comfortable continuing. 3 Month TED Spread versus 3 Month LOIS since 2010 This chart avoids the spike in 2008, making the more recent correlation between the two spreads more obvious. While they don t move exactly in line, and the TED spread tends to be more volatile (which is why I think that LOIS is the preferred indicator), they have both seen a sharp rise recently, hitting levels at or near where they have peaked since we were in the middle of the Financial Crisis. Is it Happening In Other Currencies? If this was a global funding issue, I d expect to see it appear elsewhere. LOIS for US, EU, UK and Japan since 2001 This shows that historically these LOIS spreads in various currencies have moved together. March 24,

4 During the crisis, the grey line, which is the Euro outperformed by a wide margin, and the Sterling (in pink) led the way as RBS and Barclay s were dominant in that market and had some troubles. In 2011, the Euro LOIS did worse than others, which is logical, as it was at the heart of the issue. The logical conclusion, is if this is a funding issue, it should be global. LOIS for US, EU, UK and Japan since 2010 Again, to smooth out the effect of the spike in 2008, we show the data since In 2011, when there was real concern about Europe, the EUR, USD and GBP LOIS spreads moved higher (JPY barely budged). In 2016, when money market rules went into effect, it was almost exclusively the USD libor that was impacted which makes sense again, since it was US laws that changed. The brief spike in Sterling (the pink line) in 2016 was related to Brexit again logical and largely isolated. I think we have identified that this is specifically occurring in USD which is an indication that this not global nor a true funding problem. If it was a true funding problem, I d expect borrowers to borrow anywhere they could, driving up all spreads, like we saw when there was an actual crisis. It is USD Specific But Is It Across the Curve? We have already determined that this is only occurring in U.S. LIBOR which makes me believe that this is more similar to the 2016, money market rules spike, than anything more dangerous. What we want to examine next is whether any point in the curve is being particularly hard hit which would allow us to further identify a specific cause, rather than just assuming this is an overall US Dollar funding issue. March 24,

5 1 Month, 3 Month, 6 Month and 12 Month USD LOIS Since 2010 Until now, every time spreads have moved, they have maintained a nice pattern where 1 month (in grey) is lower than 3 month (in blue) which is lower than 6 month (in pink) which is in turn lower than 12 month (in orange). The first three green ovals highlight the cross section of those spreads. Now, this time, 3 month is the obvious outlier, though 6 months isn t far behind. Why would 3 month LIBOR be doing the worst? Because that is the main maturity for bank borrowing? When I think of LIBOR, I typically think of 3 month as the one that is used most in the professional community. That is followed by 6 months. So the main maturity that I think is consistently used most by banks and corporations is doing the worst relative to others. (I view 12 month LIBOR as a retail/mortgage/home equity maturity and don t think much about 1 month though that is becoming more important as many loan documents allow the borrow to choose that maturity). Of all the LIBORS in all the world, it is USD 3 Month that is Impacted If this was a funding crisis we should see it in every currency as companies scramble for money but we don t. If this was a USD funding crisis I d expect to see it more across the curve yet we aren t we are seeing the 3 month tenor get hit particularly hard. This is strange and important. No Flight to Quality When we have seen these spreads blow out in the past there has been a flight to quality element. We will show that usually when LOIS goes higher, it is a combination of LIBOR going higher with less risky yields (OIS and T-bills) going lower. March 24,

6 3 Month LIBOR versus 3 Month OIS since 2001 During the financial crisis, the move was extreme. LIBOR shot higher while OIS plummeted. Once again, we will look at the period after the financial crisis because those numbers are so large it makes it difficult to detect more recent trends. 3 Month LIBOR versus 3 Month OIS since 2010 Here, post 2010, we see that during the four times that LOIS moved higher, it was because LIBOR moved higher while OIS moved lower or was flat (like in 2016 when it was about money market rules changing). This rise is accompanied by higher yields (3 month T-bills are also moving higher in yield) demonstrating this isn t flight to quality at all. Turning to Sherlock Holmes for Inspiration I have always liked the quote When you have eliminated the impossible, whatever remains, however improbable, must be the truth. March 24,

7 So, let s examine what we know LIBOR OIS This is not global it is USD specific, which is unusual. This is concentrated in the 3 month maturity, which is unusual. It is occurring without a flight to quality, which is unusual. Basically, we have a very isolated move, which is unusual, so let s look for what changed? Is there something that has occurred that would explain all this. What I think Happened A Simple Supply and Demand Shift Story While there are many moving parts, I am going to try and tie this to a sequence of events 1. Repatriation was passed. I think it is safe to say that much of the offshore money was being lent to out to high quality institutions for relatively short amounts of time. As those loans come due, they are being shifted into either t-bills, deposits, or being used to pay taxes or buy back stocks. Repatriation has caused some amount of money that was available for any type of loan to be smaller and have more restrictions. For every $1 billion that had been available for all types of lending, there is far less not so much because all of the money has been spent, but because it has more restrictions on what it can be invested in pulling money out of the wholesale funding market. 2. The U.S. treasury flooded the market with T-bills. As early as December 2017, reporters were questioning who would buy what is set to be the most Treasury issuance since 2010 (example). The U.S. Treasury also decided to focus on issuing T-bills rather than longer dated maturities further flooding the market with supply. 3. Asian banks have grown rapidly in the U.S. in the past several years. While many European banks pulled back in the U.S. to focus on domestic operations as the European debt crisis escalated and continued, many Asian banks Japan, China and Korea for example, have stepped up their U.S. operations. Likely an excellent business decision for them, but it means holding an increased amount of usd paper. We know for a fact that most Asian banks have March 31 st as their year-end. USD LIBOR often increases into December 31 st as U.S. banks pretty up their balance sheets for year-end reporting. Asian banks have to do similar things which increases their demand for funding (I may never have traded OIS, but I have done some massive year-end balance sheet trades with Japanese banks). Asian banks often rely on wholesale funding (U.S. banks, in contrast, have been forced to lock in lots of long-dated funds as part of various regulations, above and beyond what their competitors have had to do). Banks, as I have said before, in my experience, prefer 3 month LIBOR especially if it is a year-end balance sheet trade. So, this could explain the increased demand for USD LIBOR based funding in general and 3 months in particular. 4. The dollar has done poorly and cross currency basis swaps are expensive, so anyone holding dollar assets may want to fund those positions in dollars. After Davos, when Mnuchin basically stated that we want a weak dollar, that concern had to mount for March 24,

8 investors who had to worry about FX risk. Basically, another source of USD LIBOR based funding. 5. The Fed s balance sheet reduction, means that more money is rolling off every month than is being replaced by the Fed. That creates additional new supply that has to be bought, further reducing the availability of LIBOR type investments that are maturing and need to be replaced. I think we had a bit of a supply shock to all funding because of repatriation and the amount available for LIBOR based instruments was the most impacted. At the same time we had a demand shock as demand for short term loans increases with T-bills flooding the market and on top of that, Asian banks in particular were left scrambling for dollars ahead of year-end. What Does This Mean if I am Right? The spread should decline. LIBOR may continue to rise (along with all short term rates as the Fed still seems intent on hiking frequently), but the relative underperformance should decrease. I would expect to see the first signs of that starting in April. If I am right, and at least a portion of this spike is related to the March 31 st year-end, then we should see some easing in April. Money not currently available to the USD LIBOR market will find its way to this market. Once investors realize they are picking up 40 or 50 bps because of a rapid shift in supply and demand rather than some nefarious explanation, money will find its way to this market to close the gap. I would be finding trades that take advantage of this spread narrowing. I d also look at shorting Eurodollar futures or paying fixed on swaps. This is primarily based on the view that while LIBOR may remain elevated to other rates, that gap will close. Glen Capelo (our head of rates) and I agree that there won t be more than 3 hikes this year, which is already fully priced in. If the Fed has to tone down their hawkish rhetoric these trades should also perform well. March 24,

9 Disclaimer This document and its contents are confidential to the person(s) to whom it is delivered and should not be copied or distributed, in whole or in part, or its contents disclosed by such person(s) to any other person. Any party receiving and/or reviewing this material, in consideration therefore, agrees not to circumvent the business proposals explicitly or implicitly contained herein in any manner, directly or indirectly. Further, any recipient hereof agrees to maintain all information received in the strictest confidence and shall not disclose to any third parties any information material to the opportunity contained herein and, upon review hereof, agrees that any unauthorized disclosure by any party will result in irreparable damage for which monetary damages would be difficult or impossible to accurately determine. Recipients recognize, and hereby agree, that the proprietary information disclosed herein represents confidential and valuable proprietary information and, therefore, will not, without express prior written consent, disclose such information to any person, company, entity or other third party, unless so doing would contravene governing law or regulations. This document is an outline of matters for discussion only. This document does not constitute and should not be interpreted as advice, including legal, tax or accounting advice. This presentation includes statements that represent opinions, estimates and forecasts, which may not be realized. We believe the information provided herein is reliable, as of the date hereof, but do not warrant accuracy or completeness. In preparing these materials, we have relied upon and assumed, without independent verification, the accuracy and completeness of all information available from public sources. Nothing in this document contains a commitment from Academy to underwrite, subscribe or agent any securities or transaction; to invest in any way in any transaction or to advise related thereto or as described herein. Nothing herein imposes any obligation on Academy. Academy is a member of FINRA, SIPC and MSRB. Academy is a Certified Disabled Veteran Business Enterprise and Minority Business Enterprise, and is a Service Disabled Veteran Owned Small Business as per the US SBA. Investment Banking transactions may be executed through affiliates or other broker dealers, either under industry standard agreements or by the registration of certain principals. March 24,