Big turn by the Fed - what about the Riksbank?
Covered bonds: stable short-end conducive to carry/roll trades - buy 3Y covered bonds in ASW spreads.
Riksbank likely to wind down bond purchases.
Negative macro surprises weigh on the SEK.
Trades
New, Buy SHYP1585 ASW (Dec 2021) @ +7bp. P/L: -3bp/+15bp. 3M (NYSE:MMM) carry and rolldown:
+4bp.
Big turn by the Fed – what about the Riksbank?
Taking a step back, we have over the past couple of weeks witnessed a rather remarkable shift in tone from the world’s two leading central banks, the Federal Reserve and the ECB. The world economy is apparently slowing down. The ECB tried to stick to the ‘temporary slowdown story’ for as long as possible but at the January policy meeting it finally acknowledged that economic developments have been disappointing. In March we will see if and to what extent this perception will affect policy guidance. Fed-chair Powel, while underlining that the US economy remains healthy, specifically mentioned weakness in China and Europe as a concern along with event risks like Brexit and trade conflicts. Mr Powel confirmed that the Fed is on hold for now. We get the impression that the message is not necessarily pause and then hike later on. Instead it is a matter of pausing in order to evaluate what the next appropriate direction is; up or down.
Of course the market responded with rates going lower especially in the front-end of the US yield curve. However, considering the magnitude of the policy shift, the reaction could have been bigger had it not been for the fact that much of this was already priced in. After all, the US money market curve implies rate cuts ahead, not hikes.
The perceived recession risk has definitely risen in markets. We certainly do not question that growth is cooling off. On the other hand, economies do not just glide into a recession without a particular reason, like the Asian crises, bursting asset bubbles or other types of unsustainable imbalances working as a trigger.
A hard Brexit could surely be such a trigger. The optimistic side of the Brexit story is that no one wants a hard Brexit. The pessimistic side is that the Britons do not seem to know what they want other than a renegotiated deal, while the EU is unwilling to re-negotiate. Someone will have to blink and the clock is ticking.
In Sweden, we got a new batch (January) of consumer and business sentiment numbers this week. This time around there was a very broadly based decline across all sectors. That was not really a surprise considering that all similar survey data out of Europe are down as well but it is worth mentioning that manufacturing is holding up relatively well. Admittedly manufacturers are less upbeat on expected order flows over the coming three months but apart from that things are not that bad. This corroborates with recently presented earnings reports from companies like Volvo and Atlas Copco, which sound relatively optimistic.
Instead it is services producers (and consumers) that are gloomy. It is a mixed picture, though. Some services producers seem to be relatively unaffected, while others say that business conditions have deteriorated. The latter group include financial institutions, employment placement agencies (interesting), architects and technical consultants to mention a few. The latter probably reflect a significant decline in housing starts. Assessed labour shortage in private businesses is very high but we may have seen the peak.
The Riksbank is now in the process of figuring out what to say in the 13 February monetary policy report. Considering that several of the downside risks to the global outlook appear to be materialising, it would be a surprise to us if it does not lower the GDP projections again. Still, one could make the case that it can abstain from making any sharp policy conclusions until the April meeting when the fourth quarter GDP has been released and it has more specific knowledge about the ECB’s policy guidance. In a sense, the Riksbank is already on pause considering that the currently planned next rate hike is not until Q3. That gives the board ample time to evaluate how data evolve.
If the Riksbank lowers the GDP numbers while keeping the repo rate path unchanged, FRAs and other short-term rate instruments will almost certainly go up with the story being that the Riksbank will hike irrespective of data and projections. That would be a misinterpretation we think, contradicting the way the central bank thinks. So in our minds higher FRA rates would offer an opportunity to enter receiver positions in for instance FRA-Sep 2019.
Covered bonds: stable short-end conducive to carry/roll trades – buy 3Y covered bonds in ASW spreads
Last year was characterised by heavy supply, at least in gross terms, driven by the large redemptions. This year resembles 2017, in our view: back then we saw smaller redemptions but relatively large buybacks. If we look at the current maturity profile, we see a significant peak in 2020-21. Issuers might be keen to smooth out the maturity profile, which should speak in favour of larger buybacks this year. We expect SEK120bn, which would be slightly higher than we have seen historically.
Lending should continue to slow down. Deposit inflows remain healthy and account for an increasing share of the lending. Thus, we think that net supply will remain relatively muted. We have pencilled in SEK40bn for 2019, in line with previous years. Gross supply is set to become significantly smaller compared to previous years. The supply outlook thus appears relatively supportive.
Also interesting in our view is the spread stability in the short end, both relative to SGBs and swaps. Probably the lack of short-end material (very few T-bills, small volume in SGBs after correcting for Riksbank holdings) benefits short-dated covered bonds. This stability is clearly helpful for carry/roll-down trades further out on the curve.
Noteworthy here is that there is not a huge incentive to move far out on the curve – the carry/roll-down is relatively flat in the 2021-2024 segment, see table below.
In our view, the risks are concentrated further out on the curve. As can be seen in the chart below, the SEK covered bond curves remain very steep compared to other markets, but there is clearly a tendency for steeper curves both in the EUR and DKK market. This could at some point affect the SEK 5Y point as well.
Another risk comes from the CCS market. The relationship between SEK and EUR covered bonds (which at the margin will drive in which market mortgage institutions will prioritise issuance) is clearly linked to longer EUR/SEK CCS. Historically, the EUR/SEK CCS has correlated with what happens with EUR/USD CCS (in practice a mirror image, albeit with smaller magnitude moves in EUR/SEK CCS). Throughout 2018, USD funding has become less expensive in the EUR/USD CCS (i.e. less negative in spread terms) despite the Fed shrinking the balance sheet. There is now market speculation about the Fed being slower in shrinking the balance sheet. Although we have no strong opinion where the spread may head next, we note that EUR/USD CCS trades at levels we have not seen for years. If we see a rebound, EUR/SEK CCS is likely to move higher, which would make SEK issuance more attractive for issuers (and SEK-denominated bonds less attractive to foreign investors).
As relative-value oriented investors in our view do not really get paid more for moving further out on the curve, we prefer shorter maturities, i.e. the 2021-2022 segment, which are less exposed to a sell-off in the EUR market and issuance. Thus, we recommend buying for instance SHYP1585 in an ASW @ 7bp. We set the P/L levels to -3 bp/15bp.
ASWs in this segment have cheapened somewhat during fall.
Another development to keep track of in the housing market is the increased demand for fixed rate loans. Although there is no data where the demand is on the curve in the 1-5Y segment, we suspect that household demand is mainly concentrated in the 2-3Y segment as the lending curve out to households is relatively flat out to 3Y and somewhat steeper out to 5Y (see chart below). Generally our impression is that households tend to view fixed rate loans as insurance and are reluctant to pay up too much relative to floating rates. Banks, on the other hand, probably see an advantage with more clients having fixed rate loans (if anything lower risk and somewhat more difficult to shop around between banks) - we note that some banks have been running discounting certain fixed rate maturities.
The volumes involved are significant – over the 3M period between October-December 2018, the volume of fixed rate loans increased by SEK82bn. Although we expect that most of the loans will hit relatively liquid points on the ASW curve, if volumes continue to be large, it could put upward pressure on swap rates in the shorter end.
Riksbank likely to wind down bond purchases As mentioned the Riksbank is unlikely to continue to buy SGBs beyond the June 2019 date. Let us elaborate on this point. Although bond purchases are small relative to GDP (<7% of GDP, <8% of GDP if linkers are included), the Riksbank holds a substantial share of the outstanding volume. In the more liquid segments of the SGB curve (i.e. if we exclude SGB1056, SGB1053 and the newly introduced SGB1061), the Riksbank in most cases holds a majority of the outstanding volume. The ‘free-float’ (outstanding volume minus Riksbank holdings) is in most cases below SEK30bn. We have no doubt that the debt office is concerned about the potential harm to the liquidity and investor base.
The current bond purchases are in effect smoothed out reinvestments of SGB1052 scheduled to run until June 2019. After the maturity of SGB1052 in March 2019, the Riksbank will hold around 48% of the outstanding volume. The next nominal SGB to mature is SGB1047 in December 2020. Assuming that the latest SNDO forecast is correct (SEK30bn in 2019, SEK40bn in 2020), we have summarised how the different alternatives would affect the share of Riksbank holdings.
The extreme scenario would be if the RB decides to reinvest all holdings (SEK58bn) in SGB1047 early. Assuming the debt office forecast is correct (note that we see upside risks), the share of the Riksbank would jump to close to 57%. We reason that the Riksbank would see that as problematic.
A more likely scenario perhaps would be to reinvest portfolio coupon payments, which would be a way to wind down purchases more gradually. That would amount to around SEK8bn/year.
Overall, however, we think the need for further bond purchases can be questioned. The Riksbank is clearly on a path to normalisation of monetary policy and the sense of urgency around inflation has largely disappeared. Also, the threat of a rapidly appreciating SEK has gradually vanished. Consequently, we do not expect the Riksbank to extend bond purchases beyond June 2019 (which would keep the size of the balance sheet intact until December 2020, however). A continued reinvest of coupons is conceivable but it is not our main scenario.
Negative macro surprises weigh on the SEK
In the past couple of weeks Sweden’s macro data have in general been weaker than expected. December’s retail sales came in weaker than in years and added to disappointing household consumption for November. The yearly growth rate for household consumption has fallen from levels at 2-3% this summer to currently -0.5%, which was an unexpectedly large tumble. Earlier this week, NIER’s economic tendency survey came in weak as well, where in particular the households’ views of the economy remains pessimistic. Following all this, the SEK has shown weakness since the turn of the year and EUR/SEK is now back at levels last seen before the Riksbank’s rate hike in December, which is remarkable. The fact that data have been weaker than expected of late is evident in the surprise index, which is at its lowest levels since 2016.
It is as expected that the SEK trades down on weak Swedish macro data but, at the same time, EUR/SEK begins to look overbought in our view. We stick to our rather pessimistic view on the Swedish economy and believe that data will continue to come in weak. We are not alone in this view. On the contrary, it is widespread consensus that 2019 will turn out to be a weak year growth wise. Many forecasters, including us, hover around the Riksbank’s estimate of 1.5%. Nevertheless, this is a reason for the relatively modest strengthening of the SEK that our forecast indicates, where we see 10.20 at 1M and 10.00 at 12M. Given the recent weeks’ weak data we believe that there is a case for the markets expectations on the Swedish economy and data to adjust downwards, which limits the space for further negative surprises. By the same reasoning, this could imply a higher probability of a positive reaction in the SEK given positive, or simply neutral, macro outcomes, as the expectations have fallen. Focus can then once again shift towards the fact that the SEK seems somewhat undervalued, at least where relative rates are concerned, which strengthens our cause for a stronger SEK. Our short-term model based on relative rates yields a fair value of EUR/SEK around 10.25.
