全球-信贷策略-全球银行信贷季报：2020年展望版.docx

Global Credit Research11 December 2019US Credit ResearchKabir Caprihan, CFA AC (1-212) 834-5613kabir.x.caprihanjpmorgan J.P. Morgan Securities LLCGilead Spivack(1-212) 834-4941gilead.spivackjpmchase J.P. Morgan Securities LLCEuropean Credit Research Roberto Henriques, CFA AC (44-20)7134-1733roberto.henriquesjpmorgan J.P. Morgan Securities plcAxel J Finsterbusch, CFA(44-20)7134-4711axel.j.finsterbuschjpmorgan J.P. Morgan Securities plcDrishti Sharma (44-20)7134-0278 drishti.sharmajpmorgan J.P. Morgan Securities plc Asia Credit Research Matthew HughartAC (852) 2800-6518 matthew.hughartjpmorgan J.P. Morgan Securities (Asia Pacific) Limited Latin America Credit ResearchNatalia Corfield AC(1-212) 834-9150natalia.corfieldjpmorgan J.P. Morgan Securities LLCCEEMEA Corporate Research Konstantin Rozantsev AC (44-20) 7742-0275konstantin.rozantsevjpmorgan J.P. Morgan Securities plcGlobal Head of Credit ResearchStephen Dulake(1-212) 834-9671stephen.dulakejpmorgan J.P. Morgan Securities LLCJ. P MorganGlobal Banks Credit Quarterly2020 Outlook EditionWe relaunch our Global Banks Quarterly after a brief hiatus. This report encompasses our 2020 outlook for Asian, European, CEEMEA, North American, and Latin American banks. It has been a stellar year for financial credit spreads in general and while absolute levels look tight, there are still some pockets of value. We expect 2020 to be an interesting year, and it could very well be a tale of two halves - the current focus is on the yield environment, robust fundamental picture, and supportive technicals, but any indication of trade disruptions, a rise in geopolitical risk, and/or macro weakness will make us more cautious. On a collective basis, J.P. Morgan credit analysts are OW 17 sectors and Underweight 10. The recommendation mix reflects an improving-to-stable macro picture in DM, a lack of wide-spread bank fundamental concerns and selective preference for EM banks where fundamentals are not deteriorating. Within DM Banks, there is a clear preference for subordinate capital from European banks and Australian Tier 2 bonds over US bank bonds. We forecast 2020 global banks issuance of $61 Ibn, down 3% yoy. The largest decline is from European banks which we forecast to drop nearly 18% driven by a decline in non-preferred senior as well as holding company needs. US bank supply will rise mainly due to the refinancing of preferred stock, while EM supply will be a mixed bag as the rise in LATAM issuance is expected to be offset by a significant decline in supply from Middle Eastern and African banks. The dichotomy of issuance expectations between preferred stock and European ATI is striking - we expect US bank preferred stock issuance of about $30bn, a steep rise from 2019, while ATI is projected to drop 33% to 20bn. European ATI as well as T2 bonds from Russian and Turkish banks were the big outperformers of the year. A low yield environment, reduced geopolitical risk and decent fundamentals set the stage for financial credit to rally across the board. Within DM banks, Euro NPS, Swiss HC and UK HC p叩er were the clear outperformers and tightened 75-100bp. Within LATAM, senior Mexican bank paper tightened nearly a 140bp after a strong sell-off in late 2018 on the back of the cancellation of Mexico City's airport. Not surprising but still striking is the extent of underperformance of EM Asian Financials compared to other jurisdictions. Bank fundamentals are expected to be a mixed bag in 2020 as DM banks outperform EM - especially Asian Financials. While low and negative rates will impact bank profitability in the US and Europe, we still see that as more of an equity story. Asian Bank fundamentals will continue to deteriorate given the macro outlook for Hong Kong and China. Within LATAM, fundamentals within Brazil and Colombia should outperform other countries in the region.See page 137 for analyst certification and important disclosures.J.P. Morgan does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision. jpmorganmarkets Tier 2 Generally Outperforms in 2019U.S. subordinate holding company debt has outperformed senior holdco debt by 22bp YTD 2019, as spreads tightened 66bp to 142bp. Tier 2 issuance has remained negligible for the large US banks, which we expect will continue in 2020. We are Neutral subordinate holdco debt in 2020. The 10Y Sub/senior spread ratio of 1.2x remains well below our long term target of 1.6-1.7x, but a lack of issuance should remain supportive of spreads.European USD Tier 2 instruments performed very strongly in 2019, as tightening of 148bp outpaced tightening by US bank Tier 2 bonds and more senior parts of the European capital structure. Our European team notes that subordinated supply is near its “steady state”, and net supply is expected to be limited. This should be supportive of spreads, and the team is constructive on the space.Figure 5: Money-Center Senior vs Sub HoldcoZ Spread to Libor240100450210180150120906030908070605040302010Figure 6: European bank USD NPS vs Tier 2Asset Swap Spread10060402018016014012080400350250200150200HC Sub - HC Senior U.S. Money-Center HC Senior U.S. Money-Center HCSubSource: J.P. Morgan10050European T2 - SNP Bonds Core European Banks SNP(USD) European T2Source: J.P. MorganSupply OutlookIn our supply outlook for the European Banking sector, we highlight continued bifurcation between the Subordinated and Senior debt, with the former being at or close to 'steady state1, whereas the latter still has more to go in terms of meeting targets, in particular Bail-In Senior. We note that this narrative dovetails into our more positive view on Subordinated debt where net supply will be more limited, whereas our more cautious view on Senior debt (Underweight PrefeiTed, Neutral Bail-In) could be further challenged by continued supply. Further, we think that MREL requirements are a common denominator in European banks9 issuance plans across the capital structure. More specifically, we highlight this bifurcation as subordinated debt on the one hand approachs end-point in terms of recurrent issuers and with debut deals from smaller ones; and senior debt on the other, with several moving parts in the form of MREL and monetary policy measures continuing to breed uncertainty. We start with the simpler of the two, with past and expected ATI supply in Figure 126 below, and cumulative trends in Figure 127.AT1 SupplyOwing to a conducive market environment and a fairly heavy call schedule for the first half of 2020 (see Figure 126), ATI primary activity in 2019 was notably elevated, and set to end the year at around -31 bn gross issuance, in line with our prior 25/35bn estimates. Indeed, when compared on a cumulative basis to issuance patterns over each year since the asset class began, 2019 comes surprisingly close to retracing 2015 flows (see Figure 127). The drivers today however are entirely different - as opposed to building up their capital stacks, issuers are now nearing or have achieved their optimal stack size, and looking only to opportunistically refinance up to 6 months in advance of call dates. The predictability of issuance on this very mechanistic assumption has been almost uncanny, and gives us a fairly clear view as to where the asset class might head in the next few years. Next year, we expect issuance to take a dip towards the gross 2()bn mark, as 2H 2020 and 1H 2021 redemptions take a slight breather (see Figure 128 and Figure 129 below), and with it, expect to see net issuance finally enter the phase of "average“ net zero issuance we have been guiding towards for some time.Figure 128: AT1 Call Schedule2020DB61/4$1.25bn-43 bpING6$1bn-445bpSocGen 6$1.5bn 407bpHSBC 5 $1.5bn363bpUS5/8IBS 71/8Danbnk 5 3/41-25bn-750mn -464bj546bp DNB 5 3/4 s . BBVA$750mn-408b 61/2 63/4 Sweda 51/2 2bn S $750mn ”89bp 660bp .bpRabo51/21.5bn-)525bpLloy6 3/8750mn-EB 5 3/4529bpR$1,1bnBKIR7 3/8750mn- -385bp696bp愉:-580bp2 ANykre 61/4 1bn fsnomn-545bp -599bpBaer 8500mn - 675bpAIB 7 3/8 500mn-734bpJan-20 Feb-20 Mar-20 Apr-20 May-20 Jun-20 Jul-20 Aug-20 Sep-20 Oct-20 Nov-20 Dec-20Source: JP Morgan estimatesFigure 129: AT1 Call schedule 2021Jan-21 Feb-21 Mar-21 Apr-21 May-21 Jun-21 Jul-21 Aug-21 Sep-21 Oct-21 Nov-21 Dec-21JYBC 0SEK 1.25bn580bpRABOBK 6 5/8JYBC0DKKH.ZbDn 670bp500bn530bp3NP7 5/8 $1.5bm 631 bpSOCGEN 73/8DNBNOt)$1.5bn624bpIOK1.40 525bpnRBS8 旅NDASS 51/4$550mn 324bpACAFP )1/2$1bi 512bp$2.65bnIJBS 6 7/8 BBVASM 8 7/8 $1.5bn£ibn 918bp 550bpi760bpS IiNTAN 61/4 1.5bbn564boS juutN b 3/4HSBC>7/8 $2br551 bpDANBNK0J0.4bn 475bp k/8VMUKLN 83/4£230mn793bpISPIM7 1.25bn 688bpSHBASS 51/4 $1.2bn 334bp<00bn55 bpeMT BKTSM 8 5/8 c IPMID 8 5/8 0.20bn 125mn 836bp 887boJBS 71/8 $1.1 bn 588bpU3GIM63 1bn 610bp4ERSTBK8 0.50bn 902bpSource: J.P. Morgan estimates.However, as is always the case with bank debt, everything is not quite as simple as it seems. While on the one hand we have recurrent (and, crucially, larger) issuers approaching steady state, the conduciveness of the market this year has additionally lured in smaller institutions that so far have sat on the sidelines and chosen to rely on CETI alone to meet total capital requirements. The proportions of recurrent vs. debut issuers are shown in Figure 130 below:Figure 130: AT1 Debut vs Recurrent issuers (nominal and % of total) Debut (m) Debut (m) Recurrent (m)Source: J.P. Morgan estimates.As shown, the proportion of debut ATI issuers had stabilised somewhat at around 11% in 2017-2018, but picked up, admittedly only slightly, to 14% in 2019. Away from the obvious conduciveness of the market, we attribute this trend for debut issuers to three additional factors: 1) the impact of the negative rates environment on profitability, and therefore potential to support CETI through organic capital generation; 2) the benefit of issuing ATI in the context of impending MREL subordination requirements; and 3) Pillar 2R/2G implications, which we explain in more detail later on page 40 given the relevance for both ATI and Tier 2 issuance. Therefore, even without assuming the same degree of strength in the market next year, we would not be suiprised to see these other factors continue to draw debut issuers to the market, although we reiterate, the absolute volume is likely to be manageable and within the 5m context seen in 2019.Tier 2 SupplyMoving up the capital structure slightly, the story is in many ways not dissimilar in Tier 2, with the asset class starting to approach somewhat of a steady run rate, albeit temporarily, as Figure 131 below shows: As shown, we expect annual issuance to nbottom out" next year, with expected issuance of around 20bn, but to pick up thereafter as redemptions start to build, accumulating in a wall of c. 30bn 2013- issued 10 year bullet maturities. We must admit, we initially expected this Ubottoming-outn to materialise slightly earlier, in this year as opposed to next.Specifically, in our mid-year publication we revised down our 2019 estimates from 20bn to 15bn, following a subdued first half on a y/y basis that we attributed to lower Basel III regulatory requirements as well as the option to use cheaper senior bail-in-able instruments to meet MREL requirements. Today however, with 22bn issued in the YTD and issuers tapping the market literally as we type, the year looks due to close closer to the -25bn mark.Figure 131: Tier 2 supply (bn)Gross issuance Call scheduleSource: J.P. Morgan estimates.The strength of the market backdrop once again aside, we think the drivers here are partly shared with those we highlighted for ATI earlier, namely MREL requirements and Pillar 2G considerations, feeding through to both debut issuance and a proactiveness from recurrent issuers to build up the stack of Tier 2 often in excess of target/required levels. Indeed, both were concepts we flagged within our mid-year outlook as possible upward risks to our estimates, and where possible, we have now assigned a proportion of issuers9 indicated subordinated MREL issuance numbers to Tier 2 supply forecasts going forward. Before outlining this concept more thoroughly, however, we begin by reviewing the Pillar 2G concepts below.Upside Supply Risks: Implementing Pillar 2GAmong the various fascinating themes that have captured investors5 attention this past year, we highlight the sudden wave of debut issuers with a long list of smaller institutions bringing not only their inaugural Tier 2 transactions but also some venturing into the ATI market. Truth be told, there are multiple reasons why a bank would issue subordinated debt but we think it is worth emphasizing forthcoming changes in Pillar 2G that will create a greater incentive to fully optimize the capital structure and issue subordinated debt (both Tier 2 and ATI) so as to meet Pillar 1 requirements. We started talking about this change in our Mid-Year Outlook (page 47), but it is worth repeating the concept as we would expect changes in Pillar 2G still to be relevant in 2020 in that they will result in further debut small-sized issuers.The crux of the matter is that as of today, banks can double-count some resources and this will soon stop. Those banks meeting Pillar 1 subordinated requirements (i.e. 3.5%) with CETI, can also use that equity to meet Pillar 2G which is quite obviously double counting in that the same CETI can be use to meet both requirements (Pillar 1 and Pillar 2G) simultaneously. But that will change next year and in that way create a greater incentive to issue Tier 2 and ATI to meet Pillar 1 requirements and free-up equity to meet Pillar 2G. We think that this forthcoming change in Pillar 2G is one factor which partially explains the sudden burst of new issuers in the subordinated space and should result in further debut issuers in 2020.Figure 133: 2019 AT1 Issuance (bn)MREL SubordinationWe open this segment with some caveats, given the degree of complexity of the underlying “subordination" concept to which we committed an in-depth note earlier this year: "Seniors Only”. For the purposes of reviewing NPS and HoldCo supply, however, what is important to remember is that under BRRD2, issuers are due to receive a "subordinated" MREL requirement to complement the existi