Trillion Dollar Coin Tossing | Week of August 28, 2017

Page 1 of 39

August 25, 2017

Trillion Dollar Coin Tossing | Week of August 28, 2017 U.S. Economics: Earning Their Keep, Robert Kavcic; Michael Gregory, CFA     

Combined with results from the prior two quarters, the S&P 500 is suddenly posting some of the strongest earnings growth of the cycle, which is curious given that, by most accounts, the cycle is getting long in the tooth. Underlying economic growth has picked up in the U.S., with real GDP growth pushing back up to 2.1% y/y. The fading U.S. dollar has also removed a significant headwind. It’s no coincidence that the profit contraction reached its deepest point precisely when U.S. dollar strength was at its peak. Productivity has improved, stabilizing unit labour costs along with famously-subdued wage growth. Profit growth in technology and financials is driving the gains at the sector level, with each posting double-digit growth in Q2.

U.S. Rates: Trillion Dollar Coin Tossing, Ian Lyngen, CFA; Aaron Kohli, CFA 





In the week ahead, the market will move beyond Jackson Hole and have the benefit of incoming fundamental information to assist in further refining expectations as the unofficial end of summer looms via Labor Day weekend. The final week of August is notoriously slow in the Treasury market, with volumes and participation on the lower end of year’s range to be sure, however with core-PCE on Thursday and the BLS employment report on Friday we’re skeptical that it will be as slow as the typical dog days of summer. Let us not forget the ongoing saga in Washington that points to an increasingly contentious budget debate as the debt ceiling comes into view – perhaps Congress will once again find itself debating the merits of minting a trillion dollar coin? We certainly hope the budget/debt-ceiling issue doesn’t reach those levels again, although whose face would adorn such a coin poses an interesting thought exercise if nothing else. From a broader perspective, we remain constructive on the Treasury market and continue to anticipate that the process of cobbling together enough political support for a Continuing Resolution (CR), to keep the government functioning beyond September 30th, will further exaggerate any flight-to-quality flows that are triggered by competing headlines on the issue or elsewhere. To suggest that the situation is likely to deteriorate before there is a signed CR seems evident, at least if the experiences with healthcare repeal, deregulation, tax reforms, and infrastructure spending are any guide.

High Quality Spreads: A LIBORious Process, Dan Krieter, CFA; Dan Belton, PhD 



 

We do not believe that 2021 spells the end of Libor because the significant hurdles that stand in the way of transitioning away from Libor likely render the 2021 deadline unviable. Rather, we view last month’s announcement by the FCA as a calculated move to accomplish two objectives. First, it forces market participants with antiquated trading systems and Libor-reliant contracts to begin preparing for the transition. Secondly, it throws cold water on the thought that market participants could simply ignore the new reference rate and continue to use Libor into perpetuity. Some market participants will want to begin the transition away from Libor and to the new risk free reference rate as soon as possible, but cannot yet do so as BTFR is not yet traded. With this is mind, market participants could begin switching hedges to OIS as a more comparable rate to the future of the market than current Libor. If that happens, it may influence swap spreads higher, particularly at the long end. Our view for narrower swap spreads into year-end following a slight backup into the debt ceiling is unchanged when factoring in the impact of Libor reform; rather we expect any real impact on swap spreads from the proposed reform will not happen until 2018. The Alternative Reference Rates Committee (ARRC), chartered by the Federal Reserve, has settled on the broad Treasuries financing rate as the eventual replacement rate for Libor. While the rate is not set to be published in real time until the first half of next year, the New York Fed published historical data for the rate, giving some insight to how the new benchmark compares to other short-term rates.

Page 2 of 39

August 25, 2017

Economics: Earning Their Keep Robert Kavcic, Senior Economist; Michael Gregory, CFA, Deputy Chief Economist     

Combined with results from the prior two quarters, the S&P 500 is suddenly posting some of the strongest earnings growth of the cycle, which is curious given that, by most accounts, the cycle is getting long in the tooth. Underlying economic growth has picked up in the U.S., with real GDP growth pushing back up to 2.1% y/y. The fading U.S. dollar has also removed a significant headwind. It’s no coincidence that the profit contraction reached its deepest point precisely when U.S. dollar strength was at its peak. Productivity has improved, stabilizing unit labour costs along with famously-subdued wage growth. Profit growth in technology and financials is driving the gains at the sector level, with each posting double-digit growth in Q2.

Earning Their Keep 













Corporate earnings momentum continues to build in the U.S., with the profit recession now well in the rearview mirror. Indeed, while the S&P 500 was pushing record highs earlier this month, higher stock prices this year have largely been matched by higher earnings expectations and strong reported growth. That’s a noteworthy shift from earlier this cycle, when valuation expansion did much of the lifting for stocks. With Q2 reporting now largely in the bag, almost 74% of S&P 500 companies topped earnings expectations, according to Thomson Reuters’ tally, well above the long-run norm and the past-year average of 71%. And, it’s not just a case of the bar being lowered, as underlying earnings growth held in double-digit territory, up 12% y/y. Combined with results from the prior two quarters, the S&P 500 is suddenly posting some of the strongest earnings growth of the cycle, which is curious given that, by most accounts, the cycle is getting long in the tooth. Here are few reasons for the burst in profits: Underlying economic growth has picked up in the U.S., with real GDP growth pushing back up to 2.1% y/y from as low as 1.2% y/y in early 2016, when profits were still struggling. It’s also important that a big reason for the growth uptick has been better momentum in business investment, particularly non-residential construction and M&E—earnings growth in the industrials sector is positive again as a result. Better economic momentum in Europe has helped as well, with almost half of S&P 500 revenues derived abroad. Oil prices have helped, not only in removing a major drag from energy-sector earnings, but higher drilling activity is reinforcing the above point on business investment. Note that stripping out energy from the S&P 500 leaves earnings growth at a more modest, but still solid, 9% y/y. Additionally, the capital spending retrenchment in agriculture looks to be past the worst. The fading U.S. dollar has also removed a significant headwind. It’s no coincidence that the profit contraction reached its deepest point precisely when U.S. dollar strength was at its peak, up more than 15% y/y in September 2015. The currency has since given back nearly 10% on a trade-weighted basis, helping turn earnings momentum around. Productivity has improved, stabilizing unit labour costs along with famously-subdued wage growth. While U.S. corporate profits peaked as a share of GDP back in late-2014, likely setting a cycle high, this proxy of profit margins has at least been able to stabilize in recent quarters after compressing through mid-2016. Profit growth in technology and financials is driving the gains at the sector level, with each posting double-digit growth in Q2, and both accelerating sharply from a year ago. Banks in particular have performed well for this stage of the cycle. After persistent yield curve flattening through 2014, 2015 and much of 2016, the banks caught some relief with the 10s minus 2s curve steepening roughly 60 bps through early this year, before flattening again. The Bottom Line: Behind all the White House drama and Fed policy manoeuvring, a hearty late-cycle profit upswing is quietly playing out.

Page 3 of 39

August 25, 2017

Keys for Next Week 









Goods Trade Deficit—Monday, 8:30 am: The goods trade deficit for July will likely match June’s $64.0 bln shortfall, as narrowing and widening forces offset each other. On the narrowing side, the U.S. dollar continued to weaken (on average) in July, for the seventh consecutive month and a cumulative 5.5%. This will assist export growth along with a pick-up in global demand, particularly in Canada. On the widening side, imports should get some lift from improving domestic demand, even allowing for the dampening impact of a weaker greenback. Meanwhile, Boeing’s foreign deliveries were unchanged in the month. Real GDP—Wednesday, 8:30 am: The advance report showed real GDP expanding 2.6% annualized in Q2, more than double Q1’s clip, and subsequent data point to a net upward revision. While construction spending in June was lower than assumed, wholesale inventories were higher and there were upward revisions to retail sales. Also, the quarterly services survey revealed decent revenue growth in the quarter. On balance, we expect the second GDP growth estimate to be bumped up to 2.8%. Personal Spending and Income—Thursday, 8:30 am: With payroll employment posting a second straight 200k-plus gain and average hourly earnings rising 0.3% (a nine-month high), personal income should increase 0.4% in July after a flat June. The prior month was distorted by falling investment income as employee compensation was up 0.4%. We expect consumer spending will keep pace, reflecting the month’s strong 0.6% rise in retail sales, even allowing for lower gasoline prices. The latter will keep PCE prices to a 0.1% increase, resulting in a solid 0.3% advance in real PCE. Core PCE prices should mirror the headline move, keeping annual core inflation unchanged at 1.5% y/y and Fed concern over the lack of acceleration alive. Nonfarm Payrolls—Friday, 8:30 am: Payroll employment growth should post its third consecutive 200k-range result in August. As the improving conditions beacon more job seekers, though not enough to turn the dial on the participation rate (currently at 62.9%), the unemployment rate should remain at 4.3%. However, we see net upward risks for both metrics. With demographic forces pulling down the “part” rate, the recent 20-month run in the 62.6%-to-63.0% range is an indication of labour market strength. Average hourly earnings should increase 0.2%, lifting wage inflation a tenth to 2.6% y/y after four straight 2.5% prints and a small step along the road to finally breaking out of the sub-3% rut. Manufacturing ISM (PMI)—Friday, 10:00 am: Increasing domestic oil production continues to grease factory activity, as a weakening U.S. dollar and strengthening global demand boost the export order book. In consequence, we look for the manufacturing ISM index to increase back above the 57 level for the third time this year (shades of 2014). To date, among the regional factory surveys, most have revealed stronger results for August.

Page 4 of 39

August 25, 2017

U.S. Rates: Trillion Dollar Coin Tossing Ian Lyngen, CFA, MD, Head of US Rates Strategy, Aaron Kohli, CFA, Director, Rates Strategy Weekly Musings 







In the week ahead, the market will move beyond Jackson Hole and have the benefit of incoming fundamental information to assist in further refining expectations as the unofficial end of summer looms via Labor Day weekend. The final week of August is notoriously slow in the Treasury market, with volumes and participation on the lower end of year’s range to be sure, however with core-PCE on Thursday and the BLS employment report on Friday we’re skeptical that it will be as slow as the typical dog days of summer. Let us not forget the ongoing saga in Washington that points to an increasingly contentious budget debate as the debt ceiling comes into view – perhaps Congress will once again find itself debating the merits of minting a trillion dollar coin? We certainly hope the budget/debt-ceiling issue doesn’t reach those levels again, although whose face would adorn such a coin poses an interesting thought exercise if nothing else. More germane to the Treasury market at the moment however, is the fact that in addition to the first tier data ahead, the market will also be tasked with taking down the final round of August’s nominal supply via the 2-, 5-, and 7-year auctions. To guard against operational mistakes, the Treasury Department is auctioning both 2- and 5-years on Monday, in a move that promises to limit the set-up time and risks less non-dealer participation than one might otherwise expect. While the auction timing does make the underwriting process somewhat more challenging for the dealer community, it also means that the results will be far less influential on the overall direction of the Treasury market. Simply put, it will be much easier to explain away a series of tailed auctions at the current yield levels in light of the condensed schedule. From a broader perspective, we remain constructive on the Treasury market and continue to anticipate that the process of cobbling together enough political support for a Continuing Resolution (CR), to keep the government functioning beyond September 30th, will further exaggerate any flight-to-quality flows that are triggered by competing headlines on the issue or elsewhere. To suggest that the situation is likely to deteriorate before there is a signed CR seems evident, at least if the experiences with healthcare repeal, deregulation, tax reforms, and infrastructure spending are any guide. Recall the pledge earlier this year to have comprehensive tax reforms in place by now? We’ll suggest that as much as anything else, this is a function of the difficulty of making far-reaching structural changes that any administration would face, rather than a glib critique of this specific one. Once Congress returns from the August recess, we anticipate these tension with once again take center stage. In the interim, our attention will be on the developing economic outlook and in that context the core-PCE release will be particularly relevant. In fact, core-PCE is arguably on par with NFP in terms of importance to the market at this point, if not even more pivotal given the recent series of disappointing inflation reads. The string of five consecutive misses on core-CPI have surely scaled back expectations for PCE this week and as we contemplate a consensus +0.1% move in July and the sticker shock of a 1.4% (or below) core inflation print, we cannot help but wonder just how much inflation would need to slow before another rate hike in 2017 is taken off the table – either by the Fed or the market. On the other hand, tapering the balance sheet is still widely considered to be a foregone conclusion, although we don’t have an accurate way of measuring how much those expectations might have shifted in recent weeks. The Fed has so painstakingly telegraphed a September tapering announcement that we struggle to see anything short of a dramatic economic reversal derailing the FOMC’s intensions at this point.

Charts of the week 

Our charts of the week examine the employment data more closely with payrolls the main event in the week ahead and the market’s attention on wages as having a meaningful influence on inflation. A detailed breakdown of the wage data suggest that while inflation is tied to broader salary increases, the lack of wage pressure at the bottom of the skills spectrum highlights the hidden slack in the labor market and suggests that the flow through the Fed has counted on in the

Page 5 of 39







August 25, 2017

past may be absent for some time. The recent weakening in these measures also speaks to the fact that ostensibly “full” employment could still mean lower wage pressures and little realized inflation in the months ahead. In addition, wage gains at the bottom of the skills range were a good early indicator of past Fed cuts. A caveat is that the dispersion in wage pressures, while always present, widened in this economic cycle but now appear to be dissipating again. The latest moves in core CPI have tended to coincide more closely with lower income and wages of those over 55+ years old – suggesting that the marginal consumers (as their wages improve) might be creating more pricing pressure than those driving early-cycle improvements and who have likely already achieved a stronger/stable consumption profile as a result. The Atlanta Fed’s Current Population Survey (CPS) is a different take on distributions of wages, one that gives some insight into how the segments have experienced the latest rebound in economic growth with an alternative classification than the BLS offers on a monthly basis.

Our first chart shows the distribution of wage move the ECI, AHE and the Atlanta Fed Wage Tracker with the wide dispersion highlighting that there is a natural variance methodologies. The Atlanta Fed, for example, uses the median rather than the average figures, helping to iron out some of the variation caused by the underperformance of lower wage groups (and outperformance of higher wage earners). In recent times, the spread between the various measures has been quite large, adding to the uncertainty around the ubiquity of employment strength. For the market, wages in NFP have become a larger focus over time as inflation remains muted and the cause for such unusually weak performance is most easily blamed on still anemic wage pressures at the lower levels of the skills and earnings spectrum. The Atlanta Wage Tracker, which we examine in further detail here, offers the simplest classification for wage gains in different skillsets but is also the highest in terms of year-over-year gains of all the wage measures we routinely examine (partly due to the fact that it looks at median rather than average gains). That leaves us focusing more on the direction of the breakouts of wages than the absolute levels characterized in the Atlanta Fed data. The two categories that we examine here relate to age of workers and their skill levels. While the two offer seemingly contradictory views, they are, in fact,

Page 6 of 39

August 25, 2017

fairly consistent in painting a picture of a labor market with some pressures for skilled young workers, but absent the type of inclusive strength that has been present in past inflation and Fed hike cycles.







Our second chart shows the history of Fed rate increases against the series of wage increases broken out by skill. The first fact that the chart highlights is that while the three sets of wages by skill have tended to move together in the past, the most recent recession saw the lowest skill workers suffer larger and more persistent wage gain slowdowns than their more skilled counterparts. As such, the spread between the categories widened considerably in the latest recovery, with lower skilled workers not moving higher for several years after higher and middle skilled labor had already seen a larger bounce in wage gains. In the past, while lower skilled wages have lagged the gains of higher skilled workers, the spread has never been as wide for as long. As a result, we’ve seen the Fed respond with a lag in this cycle to improvements in the wages of lower skilled laborers rather than those of higher skilled workers, which began gaining early in the cycle (2010). The recent downtick in wages of these same lower skilled workers, while not long-lived (yet), will likely serve to dampen the Fed’s enthusiasm around the robustness or sustainability of inflation and adds to our view that the market is appropriately pricing in the lower odds of a hike later this year. The same categorization of wages, graphed against core CPI shows that CPI bounced early in 2011 on the improvement in mid and high wage workers, in what was also likely a basing effect from large deflationary impulses of 2008/2009. The latest bout of inflation gains in 2015 lagged increases in all wage groups by about a year however, it’s notable that core CPI did not show material strength until there were significant gains in the lowest skills category. In the same vein, the recent falloff in core CPI has coincided with a slowing in the wage growth of that group.

Page 7 of 39

August 25, 2017

»



A similar breakout of wages by age group shows slower growth by older groups in the workforce but much greater gains by younger workers. While that might imply faster overall growth, the fact that wages for older employees are the slowest

Page 8 of 39





August 25, 2017

likely has an outsized impact on average measures as these employees likely earn the most and incremental gains are harder to come by. By contrast, workers at the beginning of their career likely achieve more significant percentage gains as they move from lower wages to comparatively higher ones, again suggesting that their contribution to average wage gains will remain small. As such, while this explains some of the softness we’ve seen in employment markets recently, it also jibes with the results of a study by the San Francisco Fed earlier this month that younger workers may be replacing older retiring workers and pulling down average wages partly as a result. While possibly true, that does little to alter the reality of slow wage gains and our expectation that until this force is spent, we’ll see little flow through to underlying inflation. Workers who are paid at lower levels are unlikely to consume as much as baby boomers at the peak of their careers. This also says nothing of the demographic drag from retired workers who will likely also be consuming less than those with consistent employment.

Our final chart, from the San Francisco Fed and known as the wage rigidity index, measures the number of currently employed workers who report a 0% increase in wages over the prior year. The measure is a centered 12-month moving average with higher reads implying an inability to achieve wage increases. As such, the improvement by workers with some education continues the contrast of higher skilled labor achieving growth while the increasing wage rigidity of lower skilled labor speaks to some underlying softness in the job market. Thus, this latest batch of softening data in wage rigidity is not only consistent with the Atlanta Fed data, but likely adds a significant wrinkle to the Fed’s drive to view inflation pressures as transitory. With wages showing signs of softening there’s no evidence yet for the Fed’s contention either in inflation itself, or in detailed breakouts of wage gains that a low level of unemployment will turn into broader wage pressures that later turn into broader price increases.

Strategic Bias 

This week’s influences can be separated into three primary categories. First, supply considerations with the 2-, 5-, and 7year auctions leading the way early in the week, as a truncated auction schedule leaves the Treasury offerings a bit dicer

Page 9 of 39

August 25, 2017

than usual, followed by month-end extension demand. Second, the economic data that includes the risk of weaker inflation via core-PCE on Thursday followed by the BLS employment report’s NFP, UNR and AHE (of elevated importance at this particular juncture). And last, but certainly not least, the increasingly worrisome rumblings out of Washington related to the budget debt and the debt ceiling showdown. We expect supply to be more topical early in the week and the fundamentals gain in importance as the week winds down and thoughts turn to the long holiday weekend, even if there isn’t an official early close to start the celebrations on Friday. The political risks will be in place throughout the week and seem more likely to follow the news cycle as much as anything else.  While the condensed auction schedule does limit the market’s ability to set up for supply, it also makes poor results much easier to dismiss. On Monday $26 bn 2-years and $34 bn 5-years will hit the market and quickly be followed by $28 bn 7-years on Tuesday. This gross issuance figure of $88 bn is made far less daunting by offsetting maturities that total $87.5 bn and effectively leave net issuance as zero ($0.5 bn to be exact). Large maturities tend to favor the front-end auctions (2s in this case), but given the level of excess liquidity still in the market and dwindling short dated bill auctions, we struggle to imagine that there will be any major problems underwriting the new supply. Month-end extension demand will be on the upper end of the range as August was a refunding month – a dynamic that will help contain any pre-NFP downtrade.  We’ve talked a great deal about wage pressures and the different measures and what they imply in our charts of the week section, so we’ll sidestep the obvious importance of AHE in Friday’s release. It is worth noting however, that there is a strong seasonal tendency for this particularly payrolls report. Specifically, September NFP print (August’s data) tends to disappoint, having done so during each of the last six years with an average miss of -44k. A longer history (1997-2016) illustrates that the report has a 75% probability of disappointing in September and the miss is slightly larger on average at -56k. While certainly not conclusive, it is worth considering as steady jobs growth remains one of the key touchstones of the economic bulls. »

» »

Page 10 of 39 

August 25, 2017

Positions in the Treasury market continue to favor a flattener with TY and US net long as shown in the CFTC data, while the front-end of the curve remains short – albeit less so than it was just a few weeks ago. We’ll suggest that while these positions can certainly be held against a certain degree of the paintrade (i.e. bearish steepening), at the moment we’re more concerned with how limiting the positions have been in the realization of further upside. Overnight profit-taking and modest Japanese selling, seen via the MoF data (see chart), is consistent with position-squaring that we’ll argue has left the market at least marginally closer to a new equilibrium. This is also in line with our assumption that the longer 10-year yields stay in the 2.16% to 2.22% range, the larger the volume base becomes, and the easier it will ultimately prove for the market to make another move toward lower yields.

Trading View 







For our trades section this week we prefer to retain our 2s/5s steepener till our stop is reached, with the caveat that we’d pull the cord on the trade if the debt ceiling and budget impasse are resolved early. We’re not optimistic about a quick resolution when it comes to the political arena and we’d argue that remains the primary impetus for the trade, as front-end supply will be squeezed over the coming weeks. We’re also looking for steepeners on trades further up the curve, with 7s/30s and 10s/30s among the favored expressions of how we believe the curve will behave in the coming months. Historically, the 30-year has lagged rallies and selloffs with volumes in the long-bond usually low. However, recently we’ve seen that pattern reverse with bond volumes relatively higher and the long-bond outperforming on the curve. We’re not convinced that this pattern of sustained strength will continue indefinitely and we credit at least some of the gains to foreign flows, which have been fairly strong this summer (considering seasonality) after a large spate of sales to kick off the year. We’ve been focused on Japanese flows of course, but buying across the board from other investors has improved as well in recent months. We haven’t seen a great deal of sponsorship in auction stats from foreign buyers, but overnight price action has been consistently strong for the long-bond and has been a good indicator of the strength in the sector. We’ve seen the 7s/30s curve repeatedly test flatter moves over the last few weeks, with 75 bp proving a very strong support level in the charts and the technicals showing belly yields ready to break lower. The (admittedly low) negative correlation with the last year of history implies that the curve should steepen as belly yields rally. Recent moves of the curve do offer us some pause and we believe this to be a possible sign of a regime shift in the curve. While correlations are negative with a longer history, the very recent history of curve shifts and yield levels (the last six months or so) has a much lower correlation, suggesting that the curve’s response to rate moves could be in a period of transition. In the past, the curve would reflexively flatten in a selloff and steepen in a rally, while the last few months have seen the correlation turn slightly negative. We’d argue that such a pattern at the longer-end is likely to reassert itself in the coming months, as the problems of politics become more acute and possibly more intransigent (shutdown, debt-ceiling, stimulus, infrastructure etc). For now the curve’s moves haven’t been as strongly directional, perhaps responding to the dueling impulses of weaker data and political gridlock which should delay the Fed and the sporadic hopes for fiscal stimulus. »

Page 11 of 39

» »

August 25, 2017

Page 12 of 39 

August 25, 2017

As such, we’d follow the technicals on this trade for entry/exit levels, with an entry at 75 bp, just flatter than where it is now and a target at 82 bp. We’d set a tight stop at 74 bp as that has remained the bottom of a falling wedge, which also has a top around 76 bp. A break of the upper end of the triangle could well signal a much larger adjustment steeper. With this trade we’ll keep it mostly tactical, rather than focusing on fundamentals, as we observe the regime shift a little more to get a sense of which way the fundamentals and the curve will trade.

Technical Analysis 







2s – After dipping below the 50% Fibonacci retracement of the April-July selloff, the 2-year sector has drifted back toward the upper-third of the range. For support, we see the 40-day moving-average at 1.350%, which is also the beginning of a volume bulge that extends to 1.365%. Through there we see very little before a double-top at 1.394%. Momentum is skewed slightly negative with a stochastics cross that now favors higher yields. It’s notable however that stochastics are still decidedly in the middle of the range, which takes away from any implied urgency from the directional indicator. Positions in the front-end of the curve remain short as seen in the CFTC data, although roughly 60% off the recent extremes. In a rally, our first level of focus will be the yield range bottom of 1.286%, which represents significant resistance. Beyond there we’ll target the 61.8% Fibonacci retracement at 1.264%. Through those resistance levels is the 200-day moving-average at 1.252%, which is immediately followed by 1.250% -- currently the upper-end of the Fed Funds band. 5s – The price action in the belly of the curve has been consistent with the extended momentum profile and that doesn’t show any sign of deteriorating. While technically not in overbought territory, we’re cognizant that the simple fact that stochastics have spent this long near the extremes hints at a potential correction. That said, the bullish descending triangle we’ve been tracking has held and continues to point toward lower yields. We’re watching the most identifiable resistance at the bottom of the yield range from July-August at 1.726%. Beyond this, there is very little before the 1.714% low yield close for the year and then the 1.672% low yield point of 2017 achieved on June 14th. For support, we’re watching the 21-day moving-average that comes in at 1.792% and then the 40-day moving-average at 1.831%. In a more significant selloff, we have the 1.857% 200-day moving-average, but to suggest that is a stretch at this point is simply stating the obvious. We like the high closing yield of 1.827% as well. There is a volume bulge currently forming in the 1.75% to 1.77% zone that could prove a meaningful departure point for a more sustainable move. TY – Momentum in TY is arguably getting to somewhat lofty levels, although we’re reminded that through the simple process of grinding sideways these conditions can be worked off. TY has been holding an upward-sloping channel and that doesn’t appear poised to change anytime soon. For resistance, we like the 127-032 upper Bollinger band with the channel top at 127-06 beyond there. A break of the channel projects to at least the mid-June peak of 127-175. Initial support comes in at the 126-08 channel bottom and then the recent low-close of 126-035. Further weakness runs up against the 40-day moving-average at 126-00, which is also a recent low print and immediately followed by the 125-305 Bollinger bottom. Through there we see very little before the 125-16 low close. 10s – The 10-year sector is in a bullish channel that favors lower yields, although there is more room to selloff than rally if the parameters hold. The channel bottom is 2.15% and also corresponds with a volume bulge that we have been tracking and similarly provides meaningful resistance. Breaking through the high volumes zone puts into play the low yield close of 2.124% and then of course the 2.103% low yield print of 2017. A break of the range bottom doesn’t necessarily put us into uncharted territory, although we see very little before the handle-change at 2.00%. Initial support is the 21-day moving-average that comes in at 2.233% and then the 2.25% level which holds psychological significance if nothing else. Break through 2.25% and we see the 40-day moving-average at 2.271% before the channel top of 2.282%. There is also a double-top that is of some relevance at 2.283%. A further selloff would warrant a target of the 200-day moving-average at 2.338%. Momentum is still constructive and while we’re tempted to suggest it’s risking overbought, the ability of the current level in stochastics to be maintained without corrections appears to be intact.

Page 13 of 39 

August 25, 2017

30s – The long-bond has held in well with 2.75% as an important resistance level. Momentum is bullish, but as stochastics have moved into decidedly overbought territory, we’re somewhat hesitant that a correction might be in the offing. DSIs are in the middle of the range however and offer very little directional skew of any note. For resistance beyond 2.75%, we have little before the low yield-close at 2.701% and then the bottom of the yield-range at 2.682% -- a level that saw a fair amount of activity/volumes in early 2016. In a selloff, initial support can be seen at the 21-day moving-average at 2.818%. The 40-day moving-average comes in at 2.85%, which also corresponds to the recent high yield-close. Through there is something of a double-top at 2.870% and 2.874%. The 200-day moving-average is a distance away at this point and doesn’t come in until 2.958% -- which is just ahead of the 3.00% handle-change. Weekly momentum in the 30-year sector just recently shifted to favor lower yields and we expect that to be maintained into the autumn. »

»

Page 14 of 39

August 25, 2017

»

Page 15 of 39

August 25, 2017

»

Page 16 of 39

August 25, 2017

High Quality Spreads: A LIBORious Process Dan Krieter, CFA, VP; Dan Belton, PhD, Associate As summer doldrums continue to drag on, we use the time afforded from calm markets to focus on one of the largest issues facing the financial market in the years to come: Libor reform. In response to the FCA’s announcement that Libor could be discontinued beginning in 2022, ISDA released preliminary guidance on improvements to their fallback method of calculating Libor. We discuss this proposal, as well as its potential ramifications for swap spreads in the future. Additionally, we dive into the presumptive replacement rate for USD Libor, the broad Treasuries financing rate. 







We do not believe that 2021 spells the end of Libor because the significant hurdles that stand in the way of transitioning away from Libor likely render the 2021 deadline unviable. Rather, we view last month’s announcement by the FCA as a calculated move by the FCA to accomplish two objectives. First, it forces market participants with antiquated trading systems and Libor-reliant contracts to begin preparing for the transition. Secondly, it throws cold water on the thought that market participants could simply ignore the new reference rate and continue to use Libor into perpetuity. Some market participants will want to begin the transition away from Libor and to the new risk free reference rate as soon as possible, but cannot yet do so as BTFR is not yet traded. With this is mind, market participants could begin switching hedges to OIS as a more comparable rate to the future of the market than current Libor. If that happens, it may influence swap spreads higher, particularly at the long end. Our view for narrower swap spreads into year-end following a slight backup into the debt ceiling is unchanged when factoring in the impact of Libor reform; rather we expect any real impact on swap spreads from the proposed reform will not happen until 2018. The Alternative Reference Rates Committee (ARRC), chartered by the Federal Reserve, has settled on the broad Treasuries financing rate as the eventual replacement rate for Libor. While the rate is not set to be published in real time until the first half of next year, the New York Fed published historical data for the rate, giving some insight to how the new benchmark compares to other short-term rates.

High Quality Debt Underperforms Narrower Swap Spreads 

High quality spreads finish the week virtually unchanged compared to Treasuries across the curve, but underperformed narrower swap spreads 2-4bp in the 5yr sector and in (Figure 1).

vs. TSY

2 US Agency 10 Global Supra 16 European Supra 19 European Agency 31 Canadian Province 38 Covered Bond 50

vs. Swaps

3 9 16 21 29 38 57

2 3 US Agency -13 -11 Global Supra -7 -3 European Supra -4 2 European Agency 8 8 Canadian Province 16 18 Covered Bond 27 39

8/25/2017 5 7 8 9 17 17 22 15 31 35 39 42 51 --

10 21 19 20 40 47 --

2 1 0 0 1 0 -1

8/25/2017 5 7 1 13 13 21 16 19 22 37 33 45 46 --

10 25 24 25 45 52 --

2 4 3 4 4 4 1

1wk Change 3 5 7 0 -1 1 0 0 1 0 0 0 0 0 0 0 0 0 -2 -1 --

3 2 2 2 2 2 0

1wk Change 5 7 1 1 2 1 2 0 2 0 2 1 1 --

10 0 0 0 0 0 --

10 0 0 0 0 0 --

2 -3 -6 -8 -9 -5 0

2 1 0 -5 -5 -1 -2

3m Change 3 5 -1 -7 -5 -1 -6 -2 -8 -6 -3 4 -6 -7

3 1 -1 -1 -5 2 -2

7 -4 -3 -3 -3 0 --

3m Change 5 7 -3 -5 1 -2 -2 -3 -4 -3 2 -1 -6 --

10 -11 2 -6 -3 2 --

2 -0.1 -1.3 -1.6 -1.1 -0.5 -1.9

12m Z-Score 3 5 7 -0.4 -2.1 -1.6 -1.0 -1.3 -0.7 -1.4 -1.1 -1.8 -1.6 -1.3 -1.4 -1.0 -0.7 -1.1 -1.0 -1.9 --

10 -1.9 -0.7 -1.7 -0.4 -1.0 --

10 -11 2 -5 -4 1 --

2 0.7 -0.2 -0.9 -0.2 0.6 -1.2

12m Z-Score 3 5 7 0.1 -0.8 -1.1 -0.1 0.0 -0.8 -0.9 -0.9 -1.5 -1.5 -1.1 -1.3 -0.8 -0.3 -0.9 -0.2 -0.9 --

10 -1.6 -0.5 -1.2 -1.3 -0.7 --

Page 17 of 39 



August 25, 2017

The underperformance came amid a flurry of new issue activity featuring a $3 bn IBRD 3yr and $4 bn KfW 2yr. The deals were well received with very little, if any, concession to secondary market curves. This week is likely a harbinger of things to come, with September promising to be a heavy issuance month in the SSA market. This is the second week in a row that high quality spreads have underperformed swaps, reinforcing our view that spreads are likely to widen amid heavy issuance and debt ceiling related stress in September. Therefore, we make no change to the view laid out in last week’s publication, and focus on a very hot topic in today’s market: Libor reform.

Recent Developments in the Transition Away From Libor 









A few weeks ago, the U.K.’s regulatory body, the Financial Conduct Authority (FCA), announced that it would phase out Libor by year-end 2021, stating that “the market for unsecured wholesale term lending to banks…is no longer sufficiently active.” The movement towards new benchmark rates to replace Libor has been well underway across different currencies, but, although much remains uncertain, this announcement suggests for the first time that Libor may cease to be published in its current form beginning in 2022. The publishing of Libor is done through ICE Benchmark Administration (IBA), which is regulated by the FCA. The FCA has actively persuaded banks to submit Libor, and has the authority to compel panel banks to contribute. FCA CEO Andrew Bailey described the “understandable discomfort” felt by panel banks in setting Libor, explaining “we do not think it right to ask, or to require, that panel banks continue to submit expert judgements indefinitely.” This authority, granted by the European Benchmark Regulation, expires in 2020. Further, activity in the markets in which banks are charged with posting rates has declined to the point that the accuracy of the metric has been called into question. Bailey indicated in his speech that the FCA has discussed an agreement with all panel banks about voluntarily continuing to submit Libor through end-2021 in order to minimize disruption, but given the reluctance of many panel banks to contribute, there is a risk that some or most will elect to discontinue their submissions once the negotiation period ends, rendering the current format of Libor ineffective. It is important to keep in mind that the Federal Reserve chartered Alternative Reference Rates Committee (ARRC), not the FCA, is responsible for the transition from USD Libor to a new rate. Thus, FCA’s announcement presents merely a deadline on something that major regulatory bodies around the world have already been preparing for.

What Happens if Libor Disappears? 





To be clear, we do not believe that 2021 spells the end of Libor because the significant hurdles that stand in the way of transitioning away from Libor likely render the 2021 deadline unviable. Rather, we view the announcement as a calculated move by the FCA to accomplish two objectives. First, it forces market participants with antiquated trading systems and Libor-reliant contracts to begin preparing for the transition. Secondly, it throws cold water on the thought that market participants could simply ignore the new reference rate and continue to use Libor into perpetuity. It is also possible that banks continue to submit to the Libor panel following the 2021 date even though they will no longer be compelled to. After all, as banks are highly dependent on Libor in many areas of the financial system, they are incentivized, at least to an extent, to continue submitting. Still, we cannot completely discount the potential that Libor submissions will no longer be made following 2021, so we discuss what could happen if Libor publication ceases in 2021. The 2006 ISDA Definitions provide a process for determining Libor on any day that it is not published. Currently, if Libor is not published, the calculation agent is instructed to seek quotes from the London office of each reference bank. As long as at least two quotes are obtained, Libor is determined as the average of all the quotes. If that fails, the calculation agent is then to obtain quotes from the New York offices of reference banks. Again, the average of these quotes would determine Libor. There is no further guidance.

Page 18 of 39 





 





August 25, 2017

It is unclear if Libor reference banks would provide the quotes necessary to complete the fallback process if Libor is not submitted beginning in 2021, but logic says they would likely not. If the banks wanted to continue providing Libor estimates, they would likely do so in the Libor panel, allowing Libor to continue to be published. Therefore, it is nearly certain that the current fallback provisions prescribed by ISDA are insufficient to handle a permanent discontinuation of Libor. ISDA has recognized this weakness, saying in a report released last Thursday that “In most cases, fallback provisions are not sufficiently robust for a permanent discontinuation of a key IBOR.” The report can be found here. The release then goes on to discuss ISDA’s intention to improve the fallback procedures for Libor as well as some proposals for how to do so. In the “Preliminary Conclusions” section, the report outlines four potential events that would trigger the transition to a new calculation methodology for Libor. The four triggers are: o The insolvency of the relevant IBOR administrator (and there is no successor administrator) o A public statement by the relevant IBOR administrator that it will cease publishing the relevant IBOR permanently or indefinitely (and there is no successor administrator that will continue publication of the relevant IBOR) o A public statement by the supervisor for the relevant IBOR administrator that the relevant IBOR has been permanently or indefinitely discontinued o A statement by the supervisor for the relevant IBOR administrator that the relevant IBOR may no longer be used Therefore, either the FCA, as supervisor, or ICE, as administrator, has the power to trigger the fallback calculation of Libor. If this happens, ISDA put forth two potential solutions to the unsatisfactory fallback provisions. The first solution is making amendments to the ISDA 2006 Definitions to add selected fallbacks that would apply in any permanent discontinuation in the publication of Libor. These amendments would add another fallback layer to the current methodology, likely referencing newly developed benchmark rates, such as BTFR in the United States. However, these amendments would only apply to new contracts entered into after the amendments were established. Therefore, to deal with legacy contracts initiated before the amendments were put in place, ISDA is proposing a plan to amend legacy contracts referencing the applicable IBORs to include the amended definitions, including potential development of a protocol mechanism to facilitate multilateral amendments. The amendment of legacy contracts would require both parties in an existing swap agreement to agree to the amendment language; otherwise existing fallback language would remain in force. The only other option would be a mutually agreed upon termination of the swap and re-negotiation of a new one using the newly amended ISDA Definitions. Although the process would likely be onerous from a time perspective, we do not view it as a significant hurdle as market participants prepare for a world without Libor.

What Would the New Fallback Provisions Entail?  



In the above section, we discuss ISDA’s proposal for transitioning to a new rate calculation. However, likely the most important question is: what will the new rate calculation look like? ISDA provided some insight into how the rate would be developed, though significant work remains to be done. In general, ISDA proposed using alternative risk-free rates as fallbacks. That means the BTFR in the United States, SONIA for GBP Libor, and TONA for yen Libor. ISDA also acknowledged that a spread needs to be applied to the fallback riskfree rate to account for bank credit captured by the various Libors but not captured by risk free rates. ISDA outlined three key considerations when determining the spread that will be added to the risk free rate to determine Libor once one of the fallback triggers is exercised. When transitioning to the fallback calculation, there should be no opportunity to realize value from the conversion, there should be little or no potential to manipulate the spread before or after conversion, and the applicable spread should minimize distortion due to market stress at the time the fallback is applied.

Page 19 of 39 









 



August 25, 2017

Beyond that, little guidance pertaining to the actual calculation methodology of the spread is provided. However, there are a few options ISDA has at its disposal that it could use to determine the spread added to risk free rates in the fallback Libor calculation. First, ISDA could assign a fixed spread that will be added to BTFR to determine Libor once one of the fallback criteria is triggered. Most likely, this would involve taking a snapshot of the Libor and OIS market, both spot and forward, on the day Libor is discontinued, then using this snapshot to determine Libor at every future fixing over the life of outstanding contracts. As an example, consider a plain vanilla 3yr swap referencing 3m Libor initiated the day Libor ceases publication. On that day, the floating leg of the swap would be BTFR plus the 3m LIBOR/OIS spread. At the next reset in three months, the floating leg would fix as spot BTFR on that day plus the 3m forward 3m LIBOR/OIS spread determined on the snapshot date, and so on until the contract expires. A similar pattern could be applied across the curve as swaps reference various Libor tenors. This methodology seems to satisfy ISDA’s requirements of a smooth transition that offers little potential for manipulation or value realization on the day of conversion. However, there is one clear drawback of such a methodology. By locking in the spread that is added to BTFR for all future Libor fixings, Libor loses its ability to reflect funding stress at banks during times of market anxiety. Therefore, it’s conceivable that a more dynamic calculation methodology is considered, perhaps one that uses volatility or some other measure of stress in the financial system, to calculate a floating spread applied to BTFR that would more accurately reflect bank credit conditions. The obvious drawback to this possibility is the degree of complexity required to calculate the spread, as well as the higher potential for manipulation and/or gain/loss at the time of the conversion. These are just two of many possibilities that ISDA has to consider, but at this time the unknowns are too great to expand in further detail on what the new fallback calculation could mean for swap spreads both approaching and following a theoretical end of Libor. However, we do know one very important thing. No matter what spread calculation methodology is agreed upon, it will reference the BTFR. Therefore, it is vital investors become very familiar with BTFR, how it is calculated, and the questions that remain outstanding surrounding the new reference rate.

What is the Broad Treasuries Financing Rate? 







In 2014 the Fed set up the Alternative Reference Rates Committee (ARRC) with representatives from 15 big banks. The ARRC was tasked with determining an appropriate risk-free rate to eventually replace USD Libor as the rate which derivatives contracts reference. Separately, the New York Fed announced in November 2016 that it would consider publishing three benchmark rates, each based on repo transactions backed by Treasuries. The broadest of the three rates, the broad Treasuries repo financing rate (BTFR) was eventually selected by the ARRC as its preferred alternative reference rate. Specifically, the committee has focused on the overnight repo rate collateralized by Treasuries, compounded daily over a three month period. The rate is based on data from tri-party, GCF, and bilateral repo transactions. The New York Fed has proposed a methodology for trimming specials transactions from the bilateral transactional data. The New York Fed has published just less than three years of historical data for overnight BTFR, but is not set to publish the rate on a live and ongoing basis until the first half of 2018. The full implementation of BTFR as the reference rate will take years. Once the rate becomes published, the next steps include building a curve and establishing trading in BTFR futures and BTFR-OIS.

Page 20 of 39

August 25, 2017

What We Know about BTFR 



We know the general behavior of the index during the specific period in which it has been released, from November 2014 through April 2017 (Figure 2). 3 month Libor trades at a considerable spread to the secured risk-free rate, averaging around 34bp during the sample but reaching as low as 16bp and 66bp at its widest. Figure 2 is actually a geometric mean of the overnight Treasury repo rates. This, rather than the raw overnight rate, will likely be the format of the new reference rate. The overnight rate is much more volatile than the 3-month average (Figure 3), which should come as no surprise. Repo rates tend to fluctuate throughout a given month due to various supply and demand factors, and usually spike near month-end and quarter-end as dealer balance sheet becomes scarce. Figure 2: Smoothed BTFR and 3mo Libor

Figure 3: Volume-Weighted Median BTFR 100

140

90 120

80

70

100

60 bp

bp

80

50 40

60

30 40

20 20 0 Nov-14

10

Mar-15

Jul-15

Nov-15

Mar-16

Broad Treasuries financing rate

Source: BMO CM, Bloomberg. FRBNY

 



Jul-16

Nov-16

3m Libor

Mar-17

0 Aug-14

Jan-15

Jun-15

Nov-15

Apr-16

Sep-16

Feb-17

BTFR

Source: BMO CM, Bloomberg, FRBNY

It is important not to overstate the importance of the volatility depicted in Figure 2. The smoothed series are actually nearly identical regardless of whether or not the volatile quarter-end or month-end marks are excluded. OIS contracts pay a compound average of the overnight rate. We can compare both the smoothed (quarterly geometric averages) and unsmoothed OIS and BTFR rates against each other (Figure 4). Generally, Fed funds trades slightly higher than BTFR. During the end of 2016 and early 2017, the spread widened to about 12bp; this time period coincides with two rate hikes by the Federal Reserve. The spread between the two rates averaged about 3bp during the entire time period. By the time BTFR replaces Libor as the reference rate, the ARRC and the New York Fed will have developed a term structure for the rate. We expect that this will look generally like term GC rates, with liquidity highest in the overnight segment of the curve, and decreasing as the term increases. The longest standard term repo rates extend out one year, which would be relatively in line with Libor. Term GC rates are usually available in 1-, 2-, and 3-week; 1-, 2-, 3-, 6-, and 12-month tenors. The major difference between BTFR and Libor in the term structure is that overnight BTFR will be the most liquid of the new rates, while Libor’s most widely referenced term is 3 months.

Page 21 of 39

August 25, 2017

Figure 4: Fed Funds-BTFR Spread

Figure 5: BTFR Trading Volume

30

850

20

800 750

USD, billions

bp

10

0

-10

550

Mar-15

Jul-15

Nov-15

Mar-16

3-month Geometric Average

Jul-16

Nov-16

500 Aug-14

Mar-17

Jan-15

Jun-15

Nov-15

Apr-16

Sep-16

Feb-17


Overnight Raw





650

600

-20

-30 Nov-14

700

BTFR is based off of a deeply liquid market boasting an average daily trading volume of $660bn during the period it was published (Figure 5). These volumes increased over the sample, averaging $700bn during the four months of 2017 during which it was published. The major critique of Libor as a reference rate recently boiled down to it not being based on actual transactions; this should not be a significant issue with BTFR in the near-term.

How Does BTFR Compare to Other Short-Term Rates?   

Next we discuss the relationship between BTFR and other short-term rates. Figure 6 shows that interest on excess reserves usually trades higher than BTFR by an average of 15bp. BTFR tends to sit inside the Fed’s policy rate corridor, loosely bound above by interest on excess reserves (IOER) and below by the overnight reverse repo (ON RRP) facility (Figure 7). The ARRC discussed in its June 2017 meeting that BTFR trades about 4bp higher and with slightly more volatility than general collateral due to its conservative method of omitting transactions backed by collateral deemed special. Figure 6: Interest on Excess Reserves/BTFR Spread

Figure 7: BTFR Relative to Fed Policy Rates 100

40 30 80

20 10 60

bp

bp

0 -10

40

-20 -30

20

-40 -50 Aug-14

Dec-14

Apr-15

Aug-15

Dec-15

IOER-BTFR




Apr-16

Aug-16

Dec-16

Apr-17

0 Aug-14

Dec-14

Apr-15

Aug-15

Dec-15


Apr-16

ON RRP

Aug-16

Dec-16

Apr-17

IOER


Figure 8 displays the mean, standard deviations, and coefficients of variation (a normalized measure of variability obtained by dividing the standard deviation by the mean) for several pertinent short-term rates during the time period for which BTFR is available. As BTFR effectively reflects the rate of secured borrowing using a riskless asset as collateral, it is unsurprising that it posts the second lowest rate, higher only than the floor of the Fed’s policy corridor, the rate on the

Page 22 of 39

August 25, 2017

RRP facility. Perhaps more surprising, however, is the result that its standard deviation is so low, higher only than the overnight bank funding rate (OBFR) which was the other rate under consideration by the ARRC. Normalized for the lower mean, BTFR’s variability falls around the middle of the rates shown below, strikingly similar to the CVs for the raw and smoothed effective fed funds rate. That BTFR is one of the lowest rates listed below is in line with the ARRC’s mandated goal of finding a rate that was viewed as free of risk. Figure 8: Average Levels and Standard Deviations of Select Short-Term Rates (bp) Average ON RRP 21.3 BTFR (3m Geometric Avg.) 25.3 BTFR 27.3 EFFR (3m Geometric Avg.) 27.6 Effective Fed Funds 30.2 3 month OIS 34.2 Interest on Excess Reserves 44.1 1 month Treasury Bill 49.6 Overnight Bank Funding Rate 49.8 3m Libor 59.2 Source: BMO CM, Bloomberg, FRBNY *Data based on period from 8/22/14-4/28/17











Standard Coefficient Deviation of Variation 18.5 87% 15.5 61% 19.8 73% 17.6 64% 21.8 72% 23.3 68% 20.5 47% 26.0 52% 17.9 36% 30.3 51%

In addition to a sample of the rate’s behavior, we know the process by which BTFR is constructed and the data upon which it is based. The Broad Treasury financing rate will be based on three types of data: o Transaction-level data from a tri-party repo clearing platform o Activity occurring within DTCC’s General Collateral Financing (GCF) Service o Trimmed FICC-cleared bilateral Treasury repo transactions. This will not consist exclusively of GC transactions, but will be trimmed to limit the influence of specials The three sources listed above represent three repo indices that the New York Fed plans to begin to publish regularly. The first source will comprise an index known as the Narrow GC Repo Rate. The Broad GC Repo Rate will consist of the first two sources above (that is, tri-party repo and GCF data). The BTFR consists of a volume-weighted composition of all three. Given the rates of the three series released by the New York Fed, supplemented with their volumes, we can use some straightforward algebra to isolate the individual rates of the three series listed above (Figure 9). The bilateral repo rate (the one used exclusively in BTFR) is both the highest and most volatile, showing the highest propensity to spike at month-end. There are several reasons why the bilateral repo rate elevates the BTFR above the narrower GC repo rates. First, the bilateral repo transactions are the only ones containing a significant portion of specials. The process by which these specials are omitted—by dropping the lowest 25% of volume weighted transaction rates—is very conservative, and likely omits some transactions that are more likely to be considered GC. The result is an increased trimmed median. Second, bilateral repo rates are much more volatile than tri-party and GCF repo rates. This volatility in rates tends to be skewed higher, as seen in month-end and quarter-end spikes, resulting in persistently higher rates when the series is smoothed over the quarter.

Page 23 of 39

August 25, 2017

Figure 9: Data Comprising BTFR—Rates

Figure 10: Data Comprising BTFR—Volume

140

900

800

120

700 USD billions

100

bp

80 60 40

500 400 300 200 100

20

0 Aug-14

600

Jan-15 Jun-15 Tri Party Rates

Nov-15 Apr-16 Sep-16 Feb-17 GCF Rates Bilateral Repo


0 Aug-14

Jan-15

Jun-15

Tri Party Volumes

Nov-15

Apr-16

GCF Volumes

Sep-16

Feb-17

Bilateral Volumes


What Remains to be Determined? 



The most obvious and important remaining mystery about the broad treasury financing rate is the term structure. The ARRC’s goal is to construct an entire curve based on risk-free secured lending; currently, just the overnight rate has been released. While we expect that the main reference rate will be based off of the overnight rate, term rates will be utilized as well. To this point, the liquidity of BTFR’s different tenors remains to be seen. Secondly, while data for just less than three years has been released, we still know little about how the rate will perform in different market environments. For instance, we will be interested in monitoring BTFR during periods of market stress. BTFR will likely differ from Libor in this instance given that it is the rate of borrowing in a transaction secured by treasury securities. Indeed, there is a modest but positive correlation between the Libor BTFR spread and GFSI (with a correlation coefficient of 18%), a measure of market stress, but GFSI did not flare up very considerably during the period available.

What Does All This Mean for Swap Spreads? 





 



Whenever Libor is no longer published and the rate is then determined by fallback calculations based on the BTFR, there will likely be significant consequences for swap spreads. However, with the information we currently have, it is nearly impossible to determine what those consequences will be. The direction of swap spreads is heavily dependent upon the final methodology for calculating the spread that will be added to BTFR to determine Libor once fallback provisions are triggered. We do not have any clarity about what the calculation will look like at this time. Further, the term structure of BTFR and any Libor/BTFR basis that develops will be vital to estimating the impact on swap spreads. As BTFR is still not being published, remains without a term structure, and has only limited lookback data, it is essentially impossible to estimate this basis until BTFR is actively traded, both in spot and futures markets. Therefore, we await additional clarity before we can reliably estimate a likely path of swap spreads into Libor reform. However, there are some factors we look to in the meantime that could have an impact on swap spreads. Some market participants will want to begin the transition away from Libor and to the new risk free reference rate as soon as possible, but cannot yet do so as BTFR is not yet traded. With this is mind, market participants could begin switching hedges to OIS as a more comparable rate to the future of the market than current Libor. If that happens, it may influence swap spreads higher, particularly at the long end. The market participants most likely to begin switching hedges from Libor to OIS are corporate borrowers, who typically have a longer maturity issuance profile. As they do so, counterparty dealers will be incentivized to buy the OIS/Libor

Page 24 of 39





August 25, 2017

basis to hedge against any widening, which influences this basis wider. A wider OIS/Libor spread influences swap spreads wider as well. The potential for some investors to begin asset swapping positions to OIS or Treasuries instead of Libor partially mitigates the upward pressure Libor reform has on swap spreads. However, the flow is likely to be considerably smaller than issuance hedge switches, at least in the near term. Therefore, while we do not expect it to be a primary driver of swap spreads, Libor reform likely places modest upward pressure on swap spreads with accelerating pressure in the years to come. Our view for narrower swap spreads into year-end following a slight backup into the debt ceiling is unchanged when factoring in the impact of Libor reform; rather we expect any real impact on swap spreads from the proposed reform will not happen until 2018.

Past Trade Performance Figure 9: Trade Recommendations Entry Date

Trade

Benchmark

8/18/2017 8/4/2017 7/28/2017 7/7/2017 6/23/2017 6/16/2017 6/2/2017 5/19/2017 5/12/2017 5/5/2017

Overweight 3-5yr Treasuries; Underweight 3-5yr Global Supras Overweight 3yr U.S. AGY; Underweight 3yr Global Supras Overweight 10yr CAD Provincials; Underweight 10yr U.S. AGY Overweight 7yr Global Supras; Underweight 7yr Euro Supras Long 5yr Japanese SSAs 3s10s Swap Spread Flattener Overweight 5yr Euro. Agencies; Underweight 5yr Euro Supras Long 10yr European Agencies Long 5yr U.S. AGY Overweight 2yr Provincials; Underweight 2yr Global Supras

Treasuries Treasuries Swaps Swaps Swaps Treasuries Swaps Swaps Swaps Swaps

Entry Spread (bp) Last Spread (bp) -8.0 -7.5 24.1 -3.4 54.1 19.6 7.5 48.1 4.0 20.7

-11.0 -7.6 26.6 -1.8 48.7 24.8 9.0 45.0 0.1 18.5

Carry (bp/mo)

P&L (bp)

-1.2 -0.6 1.7 0.1 4.1 1.6 0.2 3.2 -0.1 2.0

2.7 -0.3 -0.9 -1.4 11.1 8.9 -1.2 10.4 3.7 2.7

Target Active/Closed Horizon 1m Active 1m Closed 1-2m Active Long-Term Active 1m Closed Long-Term Active 1m Closed 2-3m Closed 1m Closed Short-Term Closed

Exit Date

8/24/2017

8/4/2017 7/28/2017 7/28/2017 7/28/2017 5/12/2017

Source: BMO CM, TradeWeb



 



We close our recommendation from earlier this month to overweight 3yr U.S. Agencies relative to global supras taking just 0.3bp of loss over the 3 weeks. The position was initiated reflecting a bearish view on SSA spreads generally, and we saw global supra spreads widen by about 2bp during the holding period of the trade. We did not profit due to a similar widening in U.S. agencies, which we thought might be more insulated from the bearish spread environment. We maintain our 3s10s swap spread flattener which has seen substantial profits to date. We expect continued flattening as the debt ceiling approaches, but look to take profits in the next couple weeks. Our relative value trade recommendation regarding overweighting global supras relative to European supras has taken modest losses, but we maintain the position heading into the ECB’s policy decision early next month. We expect that the bank will be forced to take a more hawkish stance than it has to date, and European SSAs should underperform as a result. Additionally, we maintain last week’s recommendation to underweight global supras relative to treasuries which has taken about 3bp in profits and last month’s recommendation to overweight long-end CAD provies relative to U.S. agencies which has taken a modest loss.

Page 25 of 39

August 25, 2017

Spread Market Chart Pack High Quality Spreads to Treasuries and Swaps (bp) vs. TSY

2 US Agency 10 Global Supra 16 European Supra 19 European Agency 31 Canadian Province 38 Covered Bond 50

3 9 16 21 29 38 57

vs. Swaps

2 3 US Agency -13 -11 Global Supra -7 -3 European Supra -4 2 European Agency 8 8 Canadian Province 16 18 Covered Bond 27 39

8/25/2017 5 7 8 9 17 17 22 15 31 35 39 42 51 --

10 21 19 20 40 47 --

2 1 0 0 1 0 -1

8/25/2017 5 7 1 13 13 21 16 19 22 37 33 45 46 --

10 25 24 25 45 52 --

2 4 3 4 4 4 1

1wk Change 3 5 7 0 -1 1 0 0 1 0 0 0 0 0 0 0 0 0 -2 -1 --

3 2 2 2 2 2 0

10 0 0 0 0 0 --

1wk Change 5 7 1 1 2 1 2 0 2 0 2 1 1 --

3m Change 3 5 -1 -7 -5 -1 -6 -2 -8 -6 -3 4 -6 -7

2 -3 -6 -8 -9 -5 0

10 0 0 0 0 0 --

2 1 0 -5 -5 -1 -2

3 1 -1 -1 -5 2 -2

7 -4 -3 -3 -3 0 --

3m Change 5 7 -3 -5 1 -2 -2 -3 -4 -3 2 -1 -6 --

10 -11 2 -6 -3 2 --

2 -0.1 -1.3 -1.6 -1.1 -0.5 -1.9

12m Z-Score 3 5 7 -0.4 -2.1 -1.6 -1.0 -1.3 -0.7 -1.4 -1.1 -1.8 -1.6 -1.3 -1.4 -1.0 -0.7 -1.1 -1.0 -1.9 --

10 -1.9 -0.7 -1.7 -0.4 -1.0 --

10 -11 2 -5 -4 1 --

2 0.7 -0.2 -0.9 -0.2 0.6 -1.2

12m Z-Score 3 5 7 0.1 -0.8 -1.1 -0.1 0.0 -0.8 -0.9 -0.9 -1.5 -1.5 -1.1 -1.3 -0.8 -0.3 -0.9 -0.2 -0.9 --

10 -1.6 -0.5 -1.2 -1.3 -0.7 --

Source: TradeWeb

Swap Spreads (bp)

Swap Spreads (bp, Weekly Change) 0.0

30

-0.5 -1.0 -1.5

10

bp

bp

20

0

-2.5 -3.0

-10

-20 Mar-14

-2.0

-3.5 Sep-14

Mar-15 2yr

3yr

Source: BMO CM, Bloomberg

Sep-15 5yr

Mar-16 7yr

Sep-16 10yr

Mar-17

-4.0 2yr

3yr

Source: BMO CM, Bloomberg

5yr

7yr

10yr

Page 26 of 39

Intersector Spreads 2yr Intersector Spreads (bp) 2yr vs. Swaps US Agency Washington Supra Euro Supra Euro Agency CAD Provie Covered US Agency -6.4 9.6 21.0 29.0 40.7 Washington Supra -6.4 -3.2 14.6 22.6 34.3 Euro Supra -9.6 -3.2 -11.4 19.4 31.1 Euro Agency -21.0 -14.6 -11.4 -8.0 19.7 CAD Provie -29.0 -22.6 -19.4 -8.0 -11.7 Covered -40.7 -34.3 -31.1 -19.7 -11.7 --

2yr vs. Swaps US Agency Washington Supra Euro Supra Euro Agency CAD Provie Covered US Agency --1.1 -1.7 -1.0 -0.3 -2.0 Washington Supra 1.1 --1.4 0.0 1.1 -1.5 Euro Supra 1.7 1.4 -1.2 2.8 -0.5 Euro Agency 1.0 0.0 -1.2 -1.1 -1.3 CAD Provie 0.3 -1.1 -2.8 -1.1 --1.9 Covered 2.0 1.5 0.5 1.3 1.9 -Source: BMO CM, TradeWeb

3yr Intersector Spreads (bp) 3yr vs. Swaps US Agency Washington Supra Euro Supra Euro Agency CAD Provie Covered US Agency -7.8 12.4 18.7 29.2 49.9 Washington Supra -7.8 -4.5 10.9 21.4 42.1 Euro Supra -12.4 -4.5 -6.4 16.9 37.6 Euro Agency -18.7 -10.9 -6.4 -10.5 31.2 CAD Provie -29.2 -21.4 -16.9 -10.5 -20.7 Covered -49.9 -42.1 -37.6 -31.2 -20.7 --

3yr vs. Swaps US Agency Washington Supra Euro Supra Euro Agency CAD Provie Covered US Agency --0.5 -1.3 -1.5 -0.9 -0.5 Washington Supra 0.5 --1.7 -1.9 -1.0 -0.2 Euro Supra 1.3 1.7 --1.2 0.5 1.3 Euro Agency 1.5 1.9 1.2 -1.3 1.7 CAD Provie 0.9 1.0 -0.5 -1.3 -0.8 Covered 0.5 0.2 -1.3 -1.7 -0.8 -Source: BMO CM, TradeWeb

5yr Intersector Spreads (bp) 5yr vs. Swaps US Agency Washington Supra Euro Supra Euro Agency CAD Provie Covered US Agency -12.1 15.0 21.6 32.7 45.8 Washington Supra -12.1 -3.0 9.5 20.7 33.8 Euro Supra -15.0 -3.0 -6.6 17.7 30.8 Euro Agency -21.6 -9.5 -6.6 -11.1 24.2 CAD Provie -32.7 -20.7 -17.7 -11.1 -13.1 Covered -45.8 -33.8 -30.8 -24.2 -13.1 -5yr vs. Swaps US Agency Washington Supra Euro Supra Euro Agency CAD Provie Covered US Agency -1.8 -0.2 -0.7 0.8 -0.3 Washington Supra -1.8 --1.4 -1.7 -0.8 -1.5 Euro Supra 0.2 1.4 --1.1 0.9 -0.1 Euro Agency 0.7 1.7 1.1 -1.5 0.6 CAD Provie -0.8 0.8 -0.9 -1.5 --0.9 Covered 0.3 1.5 0.1 -0.6 0.9 -Source: BMO CM, TradeWeb

August 25, 2017

Page 27 of 39

7yr Intersector Spreads (bp)

7yr vs. Swaps US Agency Washington Supra Euro Supra Euro Agency CAD Provie US Agency -8.3 6.6 24.0 32.3 Washington Supra -8.3 --1.7 15.8 24.1 Euro Supra -6.6 1.7 -17.4 25.8 Euro Agency -24.0 -15.8 -17.4 -8.3 CAD Provie -32.3 -24.1 -25.8 -8.3 -7yr vs. Swaps US Agency Washington Supra Euro Supra Euro Agency CAD Provie US Agency -1.4 -0.6 -0.4 -0.1 Washington Supra -1.4 --1.6 -1.1 -1.0 Euro Supra 0.6 1.6 --0.1 0.3 Euro Agency 0.4 1.1 0.1 -0.3 CAD Provie 0.1 1.0 -0.3 -0.3 -Source: BMO CM, TradeWeb

10yr Intersector Spreads (bp)

10yr vs. Swaps US Agency Washington Supra Euro Supra Euro Agency CAD Provie US Agency --1.7 0.0 19.1 26.3 Washington Supra 1.7 -1.6 20.8 28.0 Euro Supra 0.0 -1.6 -19.2 26.4 Euro Agency -19.1 -20.8 -19.2 -7.2 CAD Provie -26.3 -28.0 -26.4 -7.2 --

10yr vs. Swaps US Agency Washington Supra Euro Supra Euro Agency CAD Provie US Agency -1.7 1.7 1.6 1.4 Washington Supra -1.7 --1.3 -0.3 -0.8 Euro Supra -1.7 1.3 -1.0 0.5 Euro Agency -1.6 0.3 -1.0 --0.4 CAD Provie -1.4 0.8 -0.5 0.4 -Source: BMO CM, TradeWeb

August 25, 2017

Page 28 of 39

August 25, 2017

US Agencies vs. TSY Last

1wk Chg 1m Chg 3m Chg 3m Avg 6m Avg 12m Avg 3m Z Score 6m Z Score 12m Z Score

2

9.6

0.9

2.2

-0.9

8.1

9.3

9.7

0.8

0.2

-0.1

3

8.5

0.3

0.5

0.4

8.3

8.8

9.6

0.2

-0.2

-0.4

5

7.9

-0.8

-2.4

-6.9

11.6

11.9

13.1

-1.7

-2.1

-2.1

7

8.9

0.6

-1.0

-3.0

9.8

12.6

15.4

-0.7

-1.1

-1.6

10

20.8

-0.2

-1.2

-10.8

24.4

29.7

32.4

-0.9

-1.4

-1.9

vs. TSY Last


2s3s

-1.1

-0.6

-1.7

1.3

0.2

-0.5

-0.2

-0.8

-0.4

-0.5

3s5s

-0.6

-1.1

-2.9

-7.3

3.3

3.1

3.6

-1.8

-1.7

-1.9

3s7s

0.3

0.3

-1.5

-3.3

1.5

3.7

5.9

-0.8

-1.1

-1.2

5s7s

0.9

1.4

1.4

4.0

-1.8

0.7

2.3

1.9

0.1

-0.4

7s10s

11.9

-0.8

-0.2

-7.9

14.6

17.1

16.9

-1.0

-1.6

-1.8

vs. Swaps Last


2

-13.4

3.9

4.0

-0.4

-15.6

-18.1

-16.7

0.8

1.1

0.7

3

-10.7

2.0

3.9

3.0

-12.3

-14.4

-11.5

0.8

1.2

0.1

5

0.6

1.2

1.7

-3.0

1.1

0.9

5.7

-0.2

-0.1

-0.8

7

12.9

1.0

0.1

-2.4

12.4

14.8

21.9

0.4

-0.6

-1.1

10

25.4

0.0

-0.4

-11.4

27.8

32.9

40.3

-0.6

-1.2

-1.6

vs. Swaps Last


2s3s

2.6

-1.9

-0.2

3.3

3.3

3.7

5.3

-0.3

-0.5

-1.0

3s5s

11.3

-0.8

-2.2

-6.0

13.4

15.3

17.2

-1.4

-1.7

-2.1

3s7s

23.6

-1.0

-3.8

-5.4

24.7

29.2

33.4

-0.5

-1.0

-1.5

5s7s

12.4

-0.2

-1.5

0.7

11.3

13.9

16.2

0.5

-0.4

-0.9

7s10s

12.5

-1.0

-0.5

-9.0

15.3

18.1

18.3

-1.0

-1.6

-2.0


Page 29 of 39

August 25, 2017

US Agency Technicals Gross YTD Issuance 7/31/17 $bn DNs Bullets Callables Floaters Total Non-DN Total

FNMA 338 8 6 0 14

FHLMC 224 12 41 13 66

FHLB 1,167 40 20 298 359

FFCB 103 11 11 27 48

Total 1,832 71 78 338 488

353

290

1,526

151

2,320

Source: BMO CM, FNMA, FHLMC, FHLB, FFCB Net YTD Issuance 7/31/17 $bn DNs Bullets Callables Floaters Total Non-DN Total

FNMA -7 -16 -6 -5 -26

FHLMC -11 -26 17 -9 -18

FHLB 1 -17 4 46 33

FFCB -5 -2 7 -1 5

Total -21 -60 22 31 -7

-33

-29

34

0

-27

US Agency Maturities ($ bn)

US Agency Benchmark Issuance ($ bn)

70

18 16 14

50

12

40

10 8

30

6

20

4 2

10

0 May- Dec-12 Jul-13 Feb-14 Sep-14 Apr-15 Nov-15 Jun-16 Jan-17 Aug-17 12

0 Jan

Feb Mar Apr May Jun FNMA

FHLMC

Jul

Aug Sep

FHLB

FFCB

Oct

Nov Dec

2yr

US Agency Discount Notes Outstanding ($ bn)

3yr

5yr

7yr

10yr

Source: BMO CM, Bloomberg; O:\All Strategy\Belton\Chart Pack FNMA/FHLMC Mortgage Portfolios and FHFA Sheets\Swap Spreads.xlsx

Cap ($ bn)

60%

900

50%

700

30% $ bn

% Outstanding

800

40%

20%

600 500 400

10%

300

0% Jan-11

Jan-12

Jan-13 FNMA

Jan-14 FHLMC

Jan-15

Jan-16

FHLB

Jan-17

200 Jan-08

Jan-10 FNMA

Jan-12

Jan-14 FHLMC

Jan-16

Jan-18

FHFA Cap

Source: BMO CM, Bloomberg; O:\All Strategy\Belton\Chart Pack

Source: BMO CM, FNMA, FHLMC, FHLB, FFCB Sheets\Swap Spreads.xlsx

Source: BMO CM, Bloomberg; O:\All Strategy\Belton\Chart Pack Sheets\Swap Spreads.xlsx

$ bn

$ bn

60

Page 30 of 39

August 25, 2017

Washington Supras vs. TSY Last


2

16.0

-0.3

0.4

-0.4

16.4

19.1

20.8

-0.2

-0.9

-1.3

3

16.2

0.3

0.9

0.2

15.8

17.4

20.7

0.5

-0.5

-1.0

5

16.9

0.2

-1.3

-1.6

18.0

18.8

20.6

-1.0

-0.8

-1.3

7

17.1

0.7

0.5

-2.1

17.3

19.2

19.5

-0.1

-0.7

-0.7

10

19.0

0.1

0.6

1.5

18.1

18.8

22.0

1.3

0.0

-0.7

vs. TSY Last


2s3s

0.2

0.6

0.5

0.7

-0.7

-1.7

-0.1

0.5

0.8

0.1

3s5s

0.6

-0.2

-2.2

-1.8

2.3

1.3

-0.1

-1.4

-0.3

0.2

3s7s

0.9

0.3

-0.4

-2.4

1.5

1.8

-1.2

-0.7

-0.4

0.4

5s7s

0.3

0.5

1.8

-0.5

-0.8

0.5

-1.1

0.8

-0.1

0.5

7s10s

1.9

-0.6

0.1

3.6

0.9

-0.4

2.5

0.9

0.9

-0.2

vs. Swaps Last


2

-6.9

2.8

2.3

0.1

-7.6

-9.1

-6.0

0.2

0.7

-0.2

3

-2.9

2.0

4.0

1.9

-4.5

-5.8

-2.3

0.7

1.1

-0.1

5

12.6

2.3

1.9

4.8

9.3

7.6

12.7

2.6

2.1

0.0

7

21.2

1.1

1.6

-1.8

20.0

21.1

25.1

1.2

0.0

-0.8

10

23.7

0.2

1.3

1.3

21.7

21.7

27.0

1.5

1.3

-0.5

vs. Swaps Last


2s3s

4.0

-0.8

1.8

1.8

3.1

3.2

3.7

0.8

0.7

0.2

3s5s

15.5

0.3

-2.1

2.9

13.7

13.5

15.0

0.7

1.0

0.2

3s7s

24.1

-0.9

-2.4

-3.7

24.5

26.9

27.4

-0.2

-0.9

-1.2

5s7s

8.5

-1.2

-0.3

-6.6

10.7

13.5

12.5

-1.4

-1.6

-1.4

7s10s

2.6

-0.8

-0.3

3.1

1.7

0.6

1.8

0.7

1.0

0.2


Page 31 of 39

August 25, 2017

European Supras vs. TSY Last


2

19.2

0.5

-1.0

-5.0

21.3

25.4

27.2

-1.0

-1.3

-1.6

3

20.7

0.0

-0.5

-3.0

21.9

25.2

28.1

-0.7

-0.9

-1.4

5

21.5

0.1

0.5

-1.8

22.2

26.8

29.0

-0.5

-0.7

-1.1

7

15.4

-0.1

-0.3

-7.7

18.8

23.1

24.0

-0.9

-1.4

-1.8

10

20.0

0.4

-1.9

-2.9

21.5

25.1

27.0

-1.5

-1.2

-1.7

vs. TSY Last


2s3s

1.5

-0.5

0.4

2.0

0.6

-0.2

0.9

0.8

0.6

0.2

3s5s

0.8

0.1

1.0

1.2

0.4

1.6

0.9

0.6

-0.1

0.0

3s7s

-5.3

-0.1

0.2

-4.7

-3.0

-2.1

-4.1

-0.9

-1.1

-0.3

5s7s

-6.1

-0.2

-0.8

-5.9

-3.4

-3.7

-5.0

-1.0

-0.4

-0.2

7s10s

4.6

0.5

-1.6

4.8

2.6

2.0

3.1

0.6

1.0

0.6

vs. Swaps Last


2

-3.7

3.6

0.9

-4.5

-2.4

-2.2

0.2

-0.4

-0.5

-0.9

3

1.6

1.6

2.2

-1.7

2.0

2.3

7.2

-0.1

-0.2

-0.9

5

15.6

2.2

2.7

0.2

14.6

15.9

21.5

0.6

-0.1

-0.9

7

19.5

0.3

0.3

-5.4

19.9

24.9

30.8

-0.2

-0.9

-1.5

10

25.4

0.4

0.0

-1.7

24.7

28.5

35.0

0.5

-0.7

-1.2

vs. Swaps Last


2s3s

5.3

-1.9

1.2

2.7

4.4

4.5

7.0

0.6

0.3

-0.5

3s5s

14.0

0.6

0.6

2.0

12.5

13.6

14.2

1.1

0.2

-0.1

3s7s

17.9

-1.3

-1.9

-3.7

17.9

22.6

23.5

0.0

-0.9

-1.3

5s7s

3.9

-1.9

-2.5

-5.6

5.3

9.0

9.3

-0.7

-1.2

-1.5

7s10s

5.9

0.0

-0.2

3.7

4.8

3.5

4.3

0.7

1.1

0.6


Page 32 of 39

August 25, 2017

European Agencies vs. TSY Last


2

30.5

1.0

0.6

-1.3

31.2

35.8

35.9

-0.2

-1.0

-1.1

3

28.6

-0.1

-1.5

-8.6

31.1

35.9

37.5

-1.1

-1.3

-1.6

5

30.9

0.5

-0.8

-3.3

31.4

37.6

40.7

-0.5

-0.9

-1.3

7

34.6

-0.3

-0.7

-0.8

35.3

42.7

43.4

-0.5

-1.0

-1.4

10

40.2

0.1

-0.9

-3.9

41.8

43.5

41.4

-1.1

-1.5

-0.4

vs. TSY Last


2s3s

-1.9

-1.1

-2.1

-7.3

-0.1

0.1

1.5

-0.8

-0.8

-1.0

3s5s

2.3

0.6

0.7

5.3

0.3

1.7

3.3

1.0

0.2

-0.3

3s7s

6.0

-0.2

0.8

7.8

4.2

6.8

5.9

0.9

-0.2

0.0

5s7s

3.7

-0.8

0.1

2.5

3.9

5.1

2.6

-0.1

-0.3

0.2

7s10s

5.6

0.3

-0.2

-3.1

6.4

0.8

-1.9

-0.5

0.7

1.3

vs. Swaps Last


2

7.6

4.0

2.5

-0.8

7.1

7.2

8.4

0.2

0.1

-0.2

3

8.0

2.1

0.2

-8.3

10.6

12.7

16.8

-0.8

-1.2

-1.5

5

22.2

2.4

1.4

-1.5

21.2

24.3

31.5

0.8

-0.5

-1.1

7

36.9

0.4

0.2

-1.0

36.6

42.7

48.8

0.3

-0.7

-1.3

10

44.5

0.3

0.0

-3.7

44.8

46.6

49.3

-0.2

-0.9

-1.3

vs. Swaps Last


2s3s

0.3

-1.9

-2.3

-7.5

3.4

5.5

8.3

-2.4

-1.4

-1.7

3s5s

14.2

0.4

1.2

6.8

10.6

11.5

14.7

1.3

1.0

-0.1

3s7s

29.0

-1.7

0.0

7.3

26.0

30.0

31.8

0.9

-0.2

-0.4

5s7s

14.8

-2.0

-1.1

0.5

15.4

18.5

17.1

-0.7

-0.7

-0.4

7s10s

7.6

-0.1

-0.2

-2.7

8.3

3.9

0.7

-0.5

0.5

0.9


Page 33 of 39

August 25, 2017

SSA Technicals 2017 SSA Borrowing Programs ($ bn, as of 8/3/2017)

European Supra EUR/USD Decision (bp)

European Agency EUR/USD Decision (bp)

30

30

20

20 10

10

↑ EUR

bp

0 bp

↑ EUR ↓ USD

0

↓ USD

-10

-10 -20

-20 -30

-30

-40

-40 -50 Jan-15

-50 Jan-15 Apr-15

Jul-15 2yr

Jan-16 3yr

Jul-16 5yr

7yr

Jan-17 10yr

Jul-17

Jul-15

Oct-15 Jan-16 Apr-16

2yr

3yr

5yr

Jul-16

7yr

Oct-16 Jan-17 Apr-17

10yr

Jul-17

Page 34 of 39

August 25, 2017

2017 USD SSA Maturities ($ bn)

Seasonal SSA Issuance by Month (%)

30

18% 16%

25 14%

$ bn

20

12% 10%

15 8% 10

6%

4% 5

2% 0

0% Jan

Feb Mar Apr May Jun

Jul

Aug Sep

Oct

Nov Dec

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Source: BMO CM, Bloomberg, TradeWeb

New Issues (Recent) Ticker Moody's Rating S&P Rating Announce_dt Maturity Tenor Date Month Amount (mn) JICA A+ 4/20/2017 4/27/2027 10 Year 4/20/2017 Apr 500 EBRD NR 4/25/2017 5/5/2020 3 Year 4/25/2017 Apr 1,500 ASIA Aaa AAA 4/26/2017 5/5/2020 3 Year 4/26/2017 Apr 4,000 MP Aa2 4/27/2017 5/4/2022 5 Year 4/27/2017 Apr 1,000 FMSWER Aaa AAA 5/4/2017 5/15/2020 3 Year 5/4/2017 May 1,000 EDC Aaa AAA 5/10/2017 5/17/2022 5 Year 5/10/2017 May 1,250 COE Aa1 AA+ 5/10/2017 5/17/2019 2 Year 5/10/2017 May 1,000 SEK Aa1 AA+ 5/10/2017 5/18/2020 3 Year 5/10/2017 May 1,500 ONT Aa2 5/11/2017 5/18/2022 5 Year 5/11/2017 May 2,000 KOMINS Aaa AAA 5/16/2017 6/1/2021 4 Year 5/16/2017 May 500 EIB Aaa 5/17/2017 8/14/2020 3 Year 5/17/2017 May 3,000 EIB Aaa 5/17/2017 5/24/2027 10 Year 5/17/2017 May 1,500 KFW Aaa AAA 5/22/2017 5/29/2020 3 Year 5/22/2017 May 5,000 EDC Aaa AAA 5/23/2017 6/1/2020 3 Year 5/23/2017 May 500 JBIC A1 A+ 5/24/2017 6/1/2020 3 Year 5/24/2017 May 1,500 JBIC A1 A+ 5/24/2017 6/1/2022 5 Year 5/24/2017 May 1,500 JBIC A1 A+ 5/24/2017 6/1/2027 10 Year 5/24/2017 May 1,500 ASIA Aaa AAA 5/30/2017 6/8/2021 4 Year 5/30/2017 May 1,750 KOMMUN Aaa 6/1/2017 6/12/2020 3 Year 6/1/2017 Jun 1,000 TOKYO A+ 42887 6/8/2022 5 Year 6/1/2017 Jun 500 KFW Aaa AAA 6/6/2017 12/14/2018 2 Year 6/6/2017 Jun 1,000 KBN Aaa AAA 6/13/2017 9/15/2020 3 Year 6/13/2017 Jun 1,000 SFILFR Aa3 AA 6/20/2017 6/30/2020 3 Year 6/20/2017 Jun 1,000 IADB Aaa 6/27/2017 7/7/2027 10 Year 6/27/2017 Jun 2,300 KUNTA Aa1 AA+ 7/11/2017 9/18/2020 3 Year 7/11/2017 Jul 1,000 CAF Aa3 AA7/11/2017 7/18/2020 3 Year 7/11/2017 Jul 1,250 JBIC A1 A+ 7/12/2017 7/21/2020 3 Year 7/12/2017 Jul 1,500 JBIC A1 A+ 7/13/2017 7/21/2022 5 Year 7/13/2017 Jul 1,250 JBIC A1 A+ 7/13/2017 7/21/2027 10 Year 7/13/2017 Jul 1,250 SWED Aaa 7/18/2017 7/25/2019 2 Year 7/18/2017 Jul 2,750 NRW Aa1 7/17/2017 7/25/2019 2 Year 7/17/2017 Jul 1,500 ALTA Aa1 7/19/2017 7/26/2022 5 Year 7/19/2017 Jul 1,500 FMSWER Aaa 7/25/2017 8/1/2022 5 Year 7/25/2017 Jul 1,500 KFW Aaa 7/31/2017 9/15/2021 4 Year 7/31/2017 Jul 1,000 ASIA Aaa AAA 8/1/2017 8/10/2022 5 Year 8/1/2017 Aug 750 ASIA Aaa AAA 8/1/2017 8/10/2027 10 Year 8/1/2017 Aug 500 NIB Aaa 8/2/2017 8/9/2019 2 Year 8/2/2017 Aug 500 NRWBK 8/8/2017 8/17/2020 3 Year 8/8/2017 Aug 1,000 FMSWER Aaae 8/8/2017 8/9/2019 2 Year 8/8/2017 Aug 1,000 KBN Aaa AAA 8/16/2017 11/23/2018 1 Year 8/16/2017 Aug 500 SEK Aa1 AA+ 8/22/2017 8/30/2022 5 Year 8/22/2017 Aug 1,000 IBRD Aaa AAA 8/22/2017 9/4/2020 3 Year 8/22/2017 Aug 3,000 KFW Aaae 8/22/2017 9/9/2019 2 Year 8/22/2017 Aug 4,000 DBJJP A1 A+ 8/22/2017 9/1/2022 5 Year 8/22/2017 Aug 1,000 DBJJP A1 A+ 8/22/2017 9/1/2027 10 Year 8/22/2017 Aug 800

*Green font denotes green bond

Format Launch vs MS Launch vs UST Global 61 57.1 Global -5 23.5 Global -5 23.2 Global 26 37.6 Global 0 25.5 Global 7 14.45 Global -6 19.75 Global 8 29.35 Global 29 35.6 Global 10 41.45 Global 4 26 Global 33 25.3 Global 1 21.7 Global -3 17.3 Global 48 68.5 Global 64 71.9 Global 70 64.5 Global 7 38.05 RegS 5 26.1 144A/RegS 66 73.7 Global -7 7.6 Global 8 Global 33 49.3 Global 24 21.3 Global 7 Global 45 65.6 Global 39 57.6 Global 51 57.1 Global 67 62.1 144A/RegS -6 17.3 RegS 7 30 Global 34 41.5 Global 12 19.25 Global 5 36.1 Global 9 16.3 Global 25 20.5 Global -6 18.5 Global 11 32.15 Global -4 22.7 144A/RegS -2 13.8 Global 20 Global -4 16.7 Global -4 21.5 144A/RegS 42 48.1 144A/RegS 59 53.2

Oct

Nov

Dec

Page 35 of 39

August 25, 2017

Canadian Provinces vs. TSY Last


2

38.1

0.5

3.1

0.6

35.7

39.7

40.5

1.8

-0.3

-0.5

3

38.5

0.2

-0.9

2.5

38.3

41.7

43.4

0.1

-0.6

-1.0

5

39.0

-0.4

-0.6

1.1

38.9

38.6

42.3

0.0

0.2

-0.7

7

42.1

-0.1

0.3

0.1

42.3

43.8

47.9

-0.4

-0.5

-1.1

10

47.4

-0.1

-0.7

0.7

47.2

48.7

51.9

0.3

-0.3

-1.0

vs. TSY Last


2s3s

0.4

-0.3

-4.0

1.9

2.6

2.0

2.9

-1.1

-0.6

-0.7

3s5s

0.5

-0.6

0.3

-1.4

0.6

-3.1

-1.1

-0.1

0.8

0.3

3s7s

3.6

-0.2

1.2

-2.4

4.0

2.0

4.5

-0.4

0.6

-0.2

5s7s

3.2

0.3

0.9

-1.1

3.4

5.2

5.6

-0.2

-0.8

-0.8

7s10s

5.2

0.0

-1.0

0.6

4.9

4.9

4.0

0.4

0.2

0.6

vs. Swaps Last


2

15.7

3.5

5.5

1.9

12.2

12.0

13.6

1.2

1.6

0.6

3

18.5

2.1

0.8

2.8

18.4

18.8

22.8

0.0

-0.1

-0.8

5

33.3

1.7

4.5

5.8

29.2

29.2

35.8

2.4

1.9

-0.3

7

45.3

0.5

1.1

-0.4

44.5

45.8

54.7

0.7

-0.2

-0.9

10

51.7

0.1

0.3

0.8

50.4

50.1

59.1

1.1

0.9

-0.7

vs. Swaps Last


2s3s

2.8

-1.4

-4.7

0.9

6.2

6.8

9.2

-1.4

-1.1

-1.4

3s5s

14.8

-0.4

3.7

3.0

10.8

10.4

13.0

1.8

1.9

0.4

3s7s

26.8

-1.6

0.3

-3.2

26.1

27.0

31.9

0.3

-0.1

-0.8

5s7s

11.9

-1.2

-3.4

-6.3

15.3

16.6

18.9

-2.0

-2.2

-2.2

7s10s

6.5

-0.4

-0.9

1.3

5.9

4.3

4.4

0.5

0.9

1.1


Page 36 of 39

August 25, 2017

Provincial Technicals % of FY17/18 Borrowing Program Completed

Ontario CAD/USD Funding Decision (bp)

100%

25

90%

20

80%

15

70% 10

60% bp

50%

40%

5 0

30% -5

20%

-10

10%

0% BRCOL

ALTA

SCDA

MP

ONT

Q

NBRNS

-15 Aug-16

NS

Oct-16

Dec-16 3yr

Feb-17 5yr

Covered Bonds vs. TSY Last

1wk Chg

1m Chg

3m Chg

3m Avg

6m Avg

12m Avg 3m Z Score 6m Z Score 12m Z Score

2

50.1

-1.4

-4.6

-7.6

55.6

55.9

58.5

-1.6

-1.6

-1.9

3

57.2

-1.8

0.6

-1.6

57.1

59.8

61.0

0.0

-0.6

-1.0

5

51.5

-1.3

-5.2

-5.4

56.1

59.2

60.9

-1.3

-1.4

-1.9

Last

1wk Chg

1m Chg

3m Chg

3m Avg

6m Avg

2s3s

7.1

-0.4

5.2

6.0

1.5

3.8

2.5

1.4

0.7

1.0

2s5s

1.3

0.1

-0.7

2.2

0.5

3.3

2.4

0.2

-0.4

-0.2

3s5s

-5.7

0.5

-5.8

-3.8

-1.0

-0.5

-0.1

-1.5

-1.3

-1.4

vs. TSY 12m Avg 3m Z Score 6m Z Score 12m Z Score

vs. Swaps Last

1wk Chg

1m Chg

3m Chg

3m Avg

6m Avg

2

27.4

1.5

-3.8

-7.2

32.3

30.7

12m Avg 3m Z Score 6m Z Score 12m Z Score 33.0

-1.1

-0.8

-1.2

3

39.2

-0.2

3.7

-0.1

38.1

36.8

40.3

0.3

0.8

-0.2

5

46.4

0.8

0.9

-1.2

46.4

47.3

52.6

0.0

-0.3

-0.9

Last

1wk Chg

1m Chg

3m Chg

3m Avg

6m Avg

2s3s

11.8

-1.7

7.5

7.1

5.8

6.1

7.3

1.5

1.5

1.1

2s5s

19.0

-0.7

4.7

6.0

14.1

16.6

19.6

1.2

0.5

-0.1

3s5s

7.2

1.0

-2.9

-1.1

8.3

10.5

12.3

-0.5

-0.9

-1.3

vs. Swaps


12m Avg 3m Z Score 6m Z Score 12m Z Score

7yr

Apr-17 10yr

Jun-17

Page 37 of 39

August 25, 2017

Page 38 of 39

August 25, 2017

CONTACTS Margaret Kerins, CFA, MD, Head of FI Strategy [email protected], 312-845-2687

Michael Gregory, CFA, MD, Dep. Chief Economist [email protected], 416-359-4747

Aaron Kohli, CFA, Director, FI Strategy [email protected], 212-702-1252 Dan Krieter, CFA, Vice President, FI Strategy [email protected], 312-845-4015

Sal Guatieri, Director, Sr. Economist [email protected], 416-359-5295 Jennifer Lee, Director, Sr. Economist [email protected], 416-359-4092

Dan Belton, PhD, Associate, FI Strategy [email protected], 312-845-5068

Benjamin Reitzes, Director, Canadian Rates & Macro Strategist [email protected], 416-359-5628

ANALYST’S CERTIFICATION: The authors of this report hereby certify that the views expressed in this report accurately reflect their personal views about the subject securities or issuers. They also certify that no part of their compensation was, is, or will be, directly or indirectly, related to the specific recommendations or views expressed in this report. Analysts who prepared this report are compensated based upon (among other factors) the overall profitability of BMO Capital Markets Corp. and its affiliates. Compensation for research is based on effectiveness in generating new ideas and in communication of ideas to clients, performance of recommendations, accuracy of earnings estimates and service to clients. DISCLAIMER: This material is prepared by the BMO Capital Markets Corp. (“BMOCMC”) Sales & Trading Desk, for distribution to BMOCMC clients and customers. It is not a product of the Research Department of BMOCM. The views of the BMOCMC Sales & Trading Desk may differ from those of the Research Department. This material is intended for the private use of the institutional investor to whom it was sent and is confidential and proprietary to, and may not be reproduced, disseminated or referred to, in whole or in part without the prior consent of, BMOCMC. This material has been prepared with the assistance of employees of Bank of Montreal (together with BMOCMC, “BMO”) who are involved in derivatives sales and marketing efforts. Accordingly, this material should be considered to be a solicitation of derivatives business generally and not a research report that reflects the views of disinterested research analysts. BMO is not providing advice regarding the value or advisability of trading in commodity interests, including futures contracts and commodity options, or any other activity which would cause BMO or any of its affiliates to be considered a commodity trading advisor under the U.S. Commodity Exchange Act. BMO is not undertaking to act as a swap advisor to you or in your best interests and you are urged to rely solely on advice from your qualified independent representative in making a hedging or trading decision. Notwithstanding the foregoing, this material should not be construed as an offer or the solicitation of an offer to sell or to buy or subscribe for any particular product or services (including, without limitation, any commodities, securities or other financial instruments). We are not soliciting any specific action based on this material. This material does not constitute a recommendation or a suggestion that any investment or strategy referenced herein may be suitable or appropriate for you. The information provided herein is not a complete analysis of all material facts respecting any issuer, industry or security and does not take into account the particular investment objectives, financial conditions, or needs of individual clients. Nothing in this material constitutes investment, legal, accounting or tax advice, or a representation that any investment or strategy is suitable or appropriate to your unique circumstances, or otherwise constitutes an opinion or a recommendation to you. This material is not to be relied upon in substitution for the exercise of independent judgment. You should conduct your own independent analysis of the matters referred to herein. You should seek advice based on your particular circumstances from your own independent financial, tax, legal, accounting and other professional advisors (including, without limitation, your qualified independent representative, if applicable). BMO endeavors to ensure that the information contained herein has been compiled or derived from sources that it believes are reliable and contain information and opinions which we believe are accurate and complete. However, BMO makes no representation or warranty, express or implied, in respect thereof, takes no responsibility for any errors and omissions which may be contained herein and accepts no liability whatsoever for any loss (whether direct or consequential) arising from any use of or reliance on these materials, including any information available to BMO that was not reflected herein. BMO and its affiliates may have positions (long or short), and effect transactions or make markets, in securities mentioned herein, or provide advice or loans to, or participate in the underwriting or restructuring of the obligations of, issuers and companies mentioned herein. Moreover, the BMOCMC trading desk may have acted on the basis of the information in this material. The information contained herein is as of the date and time referenced above, remains subject to change without notice although BMO does not undertake any obligation or duty to update such information. BMOCMC is a subsidiary of BMO Financial Corp., which is a subsidiary of Bank of Montreal. BMOCMC is affiliated with BMO Harris Bank NA. BMO Capital Markets is a trade name used by BMO Financial Group for the wholesale banking businesses of Bank of Montreal, BMO Harris Bank N.A. (member FDIC), Bank of Montreal Ireland p.l.c, and Bank of Montreal (China) Co. Ltd and the institutional broker dealer businesses of BMO Capital Markets Corp. (Member SIPC) in the U.S., BMO Nesbitt Burns Inc. (Member Canadian Investor Protection Fund) in Canada and Asia and BMO Capital Markets Limited (authorised and regulated by the Financial Conduct Authority) in Europe and Australia. “Nesbitt Burns” is

A daily focus on fixed income markets, trends, and trades

August 25, 2017

a registered trademark of BMO Nesbitt Burns Inc., used under license. “BMO Capital Markets” is a trademark of Bank of Montreal, used under license. "BMO (M-Bar roundel symbol)" is a registered trademark of Bank of Montreal, used under license. ® Registered trademark of Bank of Montreal in the United States, Canada and elsewhere. TM Trademark of Bank of Montreal in the United States and Canada.