SEK Views April 2018 You remain long SEK

Transcription

1 SEK Views April 2018 You remain long SEK

2 Summary SEB SEK Views survey You remain long SEK Our survey of primarily Swedish companies and institutions reveal that the local market remains surprisingly overweighted SEK (+26%), almost matching the level in the October 2017 survey at 33%. Unsurprisingly however, our polled clients at the same time expects the market now to be underweighted krona (-8%) given the relatively sharp SEK depreciation in the past months. Survey participants regard most SEK drivers as negative, and have very dovish expectations concerning the Riksbank s first repo rate hike (in-line with market pricing). So, arguably most negative news is already discounted by the market. Survey participants target jun-18 and 9.90 dec-18 in EUR/SEK. The long-term fair value for EUR/SEK is estimated at Negative interest rates are affecting corporate behaviour, with 16% stating they continue to refrain from exchanging FX into SEK. Swedish housing market outlook is expected to remain SEK negative. Most negative news seem to be discounted here. Trades: our tactically negative view on SEK has paid-off. From now we are more constructive on SEK and recommend to hedge EUR and USD exposures. Sell: EUR/SEK (FX Ringside at ). Target: Stop above Sell 8.35, target 7.50, stop on daily close above Buy 11 mths DNT in GBP/SEK vs 26%. Content 1) SEK Views survey (page 3-10) 2) Sweden: Riksbank outlook, Spring budget, Swedish portfolio flows and Current account developments (page 11-18) 3) SEB speculative SEK flows (page 19-21) 4) SEK valuation, SEBEER and relative unit labour cost development (page 22-25) 5) Riksbank rate hike cycles and historical SEK developments (page 26-30) 6) Recommendations & FX forecasts (page 31) 2

3 SEB SEK Views survey Market positioning is long SEK Our SEK Views survey of leading, mostly Swedish (but some foreign) financial institutions and companies finds that the market has kept its long SEK overweight position (+26%), compared to +33% in last October s report. With the trade-weighted krona currently at an 8-year low, it is surprising that the market remains historically long SEK. Less surprising is the fact that anticipated market positioning has fallen sharply, from +24% to -8%. Historically none of the series display any strong leading capabilities for any other, but the fact that the market is long krona but anticipates itself to be short is cause for concern given our bullish view on SEK. Compared to our survey in October 2017, financial institutions have decreased their overweight position from +62% to +25%. Meanwhile, corporates normally deviate very little from a neutral position in SEK, since by definition this must be their FX hedge benchmark. Corporates remain overweight SEK, as they were in our survey last October. 60% 40% 20% 0% -20% -40% Historical SEK positioning Overweight in SEK Underweight in SEK Own company Market 60% 40% 20% 0% -20% -40% CONCLUSION ON SEK VIEWS POSITIONING Market positioning is clearly a threat to SEK bulls Foreign accounts are obviously responsible for the SEK s weak performance (for further details, see page 19) as they are the accounts responsible for selling SEK. Participants' own SEK positioning 57.1% 35.7% 25.0% 66.7% 7.1% 8.3% Overweight Neutral Underweight 3 Corporates Financials

4 SEB SEK Views survey Market ranking of SEK drivers NEGATIVE NEWS IS CLEARLY DISCOUNTED BY NOW In our October 2017 survey we found the market to be long SEK but to also expect most krona drivers to be positive (see chart bottom right). In particular fiscal policy and exports were deemed to have underpinned the Swedish currency (and rightly so). Conversely, the Riksbank and liquidity/volatility were considered the only negative drivers, but by a narrow margin. However, in our present April 2018 survey the market remains long SEK but ranks most krona drivers as negative. Exports is still the most positive driver, which is hardly surprising given the strong global growth outlook that currently supports Swedish trade. Sweden is entering an election year and fiscal policy remains positive. Negatively, we are surprised to see such negative ranking of risk appetite, the Riksbank and Swedish home prices. SEK is little correlated to risk appetite and SEB remains constructive on risk appetite going forward. The Riksbank outlook remains dovish but the market is pricing (in-line with the survey median expectation) the first hike to occur in months time, a finding we regard as too dovish. Finally, the outlook for Swedish home prices is negative but survey respondents only count on a 2% risk premia (see page 8) from house prices to SEK, while 40% do not think house prices affect the krona at all. Furthermore SEB expects house prices to stabilise fairly soon. CONCLUSION The market seems overly pessimistic on drivers providing support for SEK 14 Apr 2018: Market ranking of SEK drivers Oct 2017: Market ranking of SEK drivers

5 SEB SEK Views survey FX hedging and Cash Management activities FX HEDGING REMAINS IN THE DOLDRUMS Since the financial crisis in 2008/09 we have asked our corporate clients if their hedging behaviour has changed. Initially, there was evidence that it had with companies decreasing the FX hedge ratio and shortening the duration of remaining hedges. In recent years, our surveys do not show any change in behaviour amongst our large corporate clients -> apparently large companies with financial Treasury departments have made few changes to their overall hedging programs. Instead we think smaller firms with less sophisticated Treasury departments (or the complete lack of one) have come to accept a larger FX exposure (less hedging activity) -> hedging has likely moved to accept more FX risk. When trying to decipher how negative interest rates have affected hedging and cash management activities we asked our respondents how they handle their cash in SEK. Despite the weak SEK, a small majority of survey participants (+16%) say they have taken action by exchanging less FX into SEK, to seek to avoid depositing cash into SEK accounts with negative interest rates. Has your FX hedge exposure changed in the past year? 24% Higher (i.e. lower FX hedge ratio) 55% Unchanged 21% Lower (i.e. higher FX hedge ratio) How has Riksbank negative rates affected your Cash Management activities 52.6% CONCLUSION Whilst the evidence is inconclusive we think companies run a higher FX exposure vs previously. Companies furthermore seem to keep some FXdenominated revenues in FX. Hence FX hedging and Cash management activities should be a small albeit positive factor for SEK. The trigger for these FX sell / buy SEK activities is rates at or above zero in Sweden. 7.9% Keep most FX-revenues 23.7% Keep some FX-revenues No change 13.2% 2.6% Hold less cashexch all FX for in FX SEK accounts 5

6 SEB SEK Views survey Weak SEK is positive but rising support to join the euro WEAK SEK IS BENEFICIAL, GROWING SUPPORT FOR JOINING THE EURO We asked survey participants what effect the current weak krona has had on their profitability. Perhaps not that surprisingly nearly 50% say exchange rate developments have been beneficial However 1 in 3 respondents state that current FX volatility is hard to manage and 1 in 8 respondents believe it is time to join the euro due to intolerable levels of exchange rate uncertainty -> a surprisingly high 45% of survey participants voice concern over the current flexible and volatile exchange rate Since clearly the current median budget rate is positive for exporters, the profit outlook for Swedish export oriented companies should be positive, taken together with slightly lower FX hedge ratios Budget rate EUR/SEK USD/SEK Average Median High Low SEK REMAINS A PRO-CYCLICAL CURRENCY BUT Compared to the last survey where only one participant said SEK is defensive, increasingly respondents believe SEK is not as dependent on the business cycle outlook as it has been previously. Nevertheless, a majority still regard SEK as a pro-cyclical currency. 47.5% 11.4% Effect of Flexible Exchange Rate Regime on Profitability 7.5% 45.5% 32.5% 27.3% 12.5% Correlation SEK and Business Cycle 9.1% 6.8% 6

7 SEB SEK Views survey No Riksbank hike until H DOVISH EXPECTATIONS ON RIKSBANK When will Riksbank hike the repo rate? Our survey respondents are in line with market pricing by forecasting a first Riksbank rate hike in H SEK remains governed by central bank policy and implicitly by inflation. Asked what factor would trigger our clients to buy SEK, their replies showed their decision to do so was clearly connected directly or indirectly to the Riksbank policy outlook 50% of respondents expect the Riksbank to hike before the ECB, so ipso facto the other 50% believe the Riksbank will raise rates after the ECB. dovish! CONCLUSION Dovish monetary policy expectations are a positive for our SEK outlook given that we forecast an earlier rate hike (Dec-18). 0.0% 2.2% 8.9% 8.9% 15.6% 28.9% 20.0% 15.6% 41.2% Factors triggering SEK purchases 29.4% 11.8% 5.9% 2.9% 2.9% 2.9% 2.9% 7

8 SEB SEK Views survey Dovish expectations on Swedish housing market and krona POOR OUTLOOK FOR THE SWEDISH HOUSING MARKET The median respondent expects Swedish house prices to continue to decline over the next 12 months by a further 3% The median respondent thinks that EUR/SEK carries a 2.4% risk premium to reflect the Swedish housing market outlook (i.e. EUR/SEK trades 2.4% higher given recent housing market developments) Regarding transmission of the house price decline into SEK, 6 out of 10 expect a weaker SEK as the Riksbank delays interest rate hikes or foreign investors sell Swedish assets as a result of the house price decline 27.9% House price decline impact on the SEK? 39.5% 32.6% 0.0% NOK - THE FUTURE STARPERFORMER Survey respondents rank NOK as the clear winner in the coming 6 months. We agree EUR/NOK will head lower but think the potential in SEK is equally strong On a separate question we asked our Corporate clients to rank the currency of most importance to them. EUR (64%) and USD (34%) were the two sole answers CONCLUSION We have estimated the SEK risk premium from the housing market at 1%. The market as a whole is more negative and 60% of respondents believe house prices matter negatively for SEK. So far, the housing market s overall macroeconomic implications have been small. Again the surprise factor here is SEK positive on a stabilising outlook for Swedish house prices later this year. Which currency will outperform all others over the next 6 months? 50% 16% 20% 14% Euro Dollar Norwegian krone Swedish krona 8

9 SEB: SEK and the housing market risk premium We still estimate a small risk premium of % THE POSITIVE SEK IMPACT OF A STABILIZING HOUSING MARKET IS LIMITED Our indicator of the risk premium in SEK due to the housing market still suggests there has been little impact on the currency from falling house prices. However, we may have slightly underestimated the risk premium, as we use the rate spread to compare EUR/SEK with and without a housing market impact: Base model: 5y rate spread to explain EUR/SEK Augmented model: 5y rate spread and OMX real estate index Potentially, the 5y rate spread might also have been affected by concerns over the housing market (resulting in lower rates than otherwise) but this effect will not show up in our indicator. Therefore, a risk premium of between 0.6% and 1.2% seems more appropriate, although it is still lower than the average estimate in the SEK survey. 9

10 SEB SEK Views survey Krona forecasts and estimated long-term fair values for SEK SEK IS UNDERVALUED AND SHOULD APPRECIATE IN 2018 The current weak SEK exchange rate is not expected to last: the median respondent expects EUR/SEK and USD/SEK to fall by 4-5% to year-end 2018 (targeting EUR/USD flat at 1.23 by Dec-18). When asked about the estimated long-term fair value for SEK, the median respondent targets 9.30 and 7.60 respectively in EUR/SEK and USD/SEK (giving a EUR/USD fair-value at 1.22). Our own SEB long-term fair value model SEBEER puts EUR/SEK and USD/SEK fair values at 9.13 and Please also see pages where we discuss Swedish domestic unit labour cost developments over the past 20 years. These are indicative of SEK being less undervalued currently vs. our more traditional equilibrium models. In our SEK Views report last October, we asked respondents what they considered the probability that Sweden will join the euro by 2025 (10%). Our current survey finds the likelihood has increased to an average of 20%. While we see no strong fundamental reason why this development has occurred, we believe Sweden is moving closer to euro membership. More on page 6. The survey finds the estimated accession rate at which Sweden would fix the SEK exchange rate to the euro to be This rate is also very similar to respondents EUR/SEK long-term fair value (9.30). SUMMARY SEK PROJECTIONS As is almost always the case, the median respondent forecasts a stronger SEK. Whilst positioning is a headwind, valuation clearly supports such a projection for the next 6-12 months. EUR/SEK EUR/SEK USD/SEK USD/SEK Jun-18 Dec-18 Jun-18 Dec-18 Average Median High Low Fair-value EUR/SEK USD/SEK Average Median High Low Prob (%) Swe to join the euro Average 20.3 Median 17.5 High 70.0 Low 0.0 Eur/sek accession rate Average 9.26 Median 9.25 High Low

11 Sweden: Lower service inflation delays 1st hike Riksbank expected to lower its rate path in April SUBDUED SERVICE AND WAGE INFLATION TO POSTPONE FIRST RATE HIKE Due to higher electricity prices and weaker krona, CPIF is predicted to be close to the target and slightly above the Riksbank s estimate throughout the year. However, service inflation has turned markedly lower since last year s peak and together with muted wage inflation creates a headache for the Riksbank. CFIF ex energy was approximately 0.2%-points below the Riksbank s forecast in January and February and is expected to remain lower also going forward. With most Riksbank board members still very focused on spot inflation and increasing concerns for service and wage inflation, we predict the Riksbank will postpone its indicated first hike from September at its April meeting. As we see risks for service inflation to decline further, we also continue to see downside risks to our forecast of a first Riksbank hike in December. After having been largely in line with the Riksbank s repo rate path, market expectations have shifted markedly lower since mid-february and current pricing suggests 18bps hike in Dec-18 while a full 25bps hike is priced in Feb-19. Although we regard current market pricing as relatively fair given our Riksbank s expectations, we see a risk that if March CPIF is in line with our forecast of 2.1%, the Riksbank may be unwilling to make any larger changes in its path in April. Hence, this would be regarded as a hawkish signal and probably shift market expectations temporarily slightly higher again. However, in general we expect the Riksbank s policy to support short rates to stay low over coming months while longer rates are expected to follow international rates slightly higher. 11

12 Sweden: Spring budget In election-waiting-mode STRONG GOVERNMENT FINANCES Government finances have been in surplus for the last 3 years. Central government debt is at the lowest level since the 1970ies, below 30% of GDP (General government debt approximately 10pp higher). Continuous positive fiscal surprises in the last years as growth has been driven by high-taxed areas such as construction and consumption. Also, rising property and equity prices have pushed up capital gains tax at the same time as employment has increased at a fast pace. As growth edges down and industry/exports become more of a driving factor, tax revenue growth will also be at a slower pace. Revised fiscal policy rule from 2019: 1/3% of GDP government surplus over the cycle and a debt anchor at 35% of GDP. As debt is heading towards the anchor the issue of what to adjust might up for discussion earlier than expected. KRONA-FOR-KRONA REPLACED BY LATE CYCLE SPENDING SPREE Strong business cycle and government finances in good shape has created room for an expansionary election budget Fiscal policy was largely neutral in , turned expansionary in Like Trumps US tax cuts, timing is not optimal. Fiscal stance will be an issue in the election campaign but still difficult to criticise government as the Riksbank continues with expansionary policy. We expect a spring budget with only minor new proposals. Normally the Spring Budget is the starting point for the budget preparations for the next year where the government presents major areas of reform, with details to come in the autumn Budget Bill. As we are in an election year, the government will not show its full hands. Parties election programmes are usually presented in pieces during summer/early autumn. 12

13 Sweden: Portfolio flows Net equity transactions recorded an outflow of SEK 125bn in 2017 SWEDISH APPETITE FOR FOREIGN EQUITIES A DECIDING SEK FACTOR?! We regard Swedish net equity flows as highly relevant for the SEK development During H Swedish investors net sold SEK 100bn of foreign equities contributing to a stronger krona during that period On Jan 4, 2016, the Riksbank responded to a strong krona by adopting the FX intervention mandate for Stefan Ingves and Kerstin af Jochnick, hence Riksbank was partly fighting structural factors for the exchange rate During 2017, equities showed a steady net outflow, i.e. Swedish investors net bought foreign equities was also a period characterized by positive global risk appetite and rising stock markets. Therefore, once again we think it fair to say that in times of rising risk appetite, Swedish investors tend to increase their overseas equity exposure. This also means that during times of declining risk appetite the contrary should hold. In particular, the Swedish net investment position in equities is very positive, i.e. Sweden owns more equities abroad vs. foreign investors in Swedish equities. Consequently, from this perspective SEK now trades defensively rather than with a pro-cyclical bias (which many still believe is the predominant feature of the krona s trading pattern). As regards foreign ownership of Swedish equities, the share actually increased to almost 41% Swedish government bonds continue to see a decline since the introduction of quantitative easing by Riksbank (now below 25%). CONCLUSION: Given 1) large positive net equity position and 2) limited foreign ownership of Swedish bonds, portfolio flows have a skew to be SEK positive

14 Sweden: Current account Long downward trend for current account surplus GRADUAL DECLINE IN CURRENT ACCOUNT AND NET TRADE AFTER THE FINANCIAL CRISIS The Swedish current account surplus continues to be large (on average 3.2% of GDP in 2017), as share of GDP the surplus has, however, trended gradually lower since a peaking at 8.5% before the financial crisis The driving force for the decline is a gradual decline for net trade with goods and services which started in The downward trend is driven by a gradual decline in the surplus for goods, which at the end of 2016 turned marginally negative. The downward trend in net goods trade is a driven partly by a weak international recovery after the financial crisis, but more important is a strong trend for imports due to strong domestic demand, which has been strong both in absolute terms and relative other countries. Parts of the downward trend for goods has been offset by an upward trend for net trade with services, which has increased from slightly above 0.5% in 2007, to 2% in Equally important for the Swedish surplus is merchanting, which is defined as income generated by trade with goods that does not cross the Swedish boarder. This income is generated by the many international Swedish companies with a large share of their production in other countries, but with administration, research and development and sales organisations in Sweden. Merchanting explains 2%-points of the Swedish trade surplus, which is approximately twice as large as the corresponding share for e.g. Germany. 14

15 Sweden: Current account Expected to stabilise in 2018/19 CURRENT ACCOUNT EXPECTED TO STABILISE IN 2018/2019 Other income explains 1.5%-points of the decline in the current account surplus since 2007, although the decline has been halted over the last 2-3 years. A decline in net investment income (net capital income) is the most important driver of the downturn. Net investment income is still positive, but a negative contribution from net contribution to EU and foreign aid means that the net current account payments excluding trade is slightly negative. Despite the fact that the decline in the current account surplus accelerated in H2 2017, there are reasons to expect the surplus to level out or even increase slightly in 2018 and The first and most important reason is that exports are expected to accelerate due to the stronger international business cycle. This can be seen in that the downward trend for goods trade was halted in the middle of last year and net goods trade over the last 6-12 months is showing signs of increasing. Another reason is that the strong upturn in domestic demand can be exported to ease, with more specifically the recent strong increase in residential investments is predicted to turn into a marked decline in A weaker krona is also likely to be supportive for the trade balance. Normally the exchange rate affects the trade balance with a lag (J-curve effect) with a weaker exchange rate initially leading to a slightly weaker trade balance through higher prices, but with volume effects dominating on the slightly longer term. 15

16 Sweden: Current account Higher investments, lower financial savings UNUSUAL WITH LARGE FINANCIAL SAVINGS AND HIGH INVESTMENTS The fast decline in the surplus in 2017 is an uncertain factor. The lower trade surplus last year was explained by a marked decline in the net service trade, with both weak exports and very strong import of services contributing to the decline. The weaker exports is in our view a reaction to a strong upturn in 2015 and we expect the upward trend to be resumed this year. The strong imports was driven by increased foreign travels and financial services. Given the weak krona it seems unlikely that the upturn will continue while the large increase in financial services looks very much as a one-off increase in the beginning of last year. It is also worth observing that the service trade statistics is often revised and that the large negative contribution from service trade is the most important reason why GDP in 2017 looks mysteriously weak compared to sentiment indicators and the labour market. An alternative way to look at the current account balance is to look at financial savings in different sectors. By definition the current account is the sum of the financial savings in the corporate, government and household sectors. The downward trend in the current account over the last 3-4 years is almost totally explained by lower savings among non-financial corporations, while household financial savings have remained stable on a high level and government savings has improved. The lower financial savings in the corporate sector is explained by higher investments, with the investment as share of GDP having increased to the highest level since the mid 1990s during Higher residential investments is the main driver of the rising investments relative to GDP, while other investments have been stable on a moderately high level. 16

17 Sweden: Current account Total national savings have increased FIXED INVESTMENTS HAVE INCREASED MORE THAN THE DECLINE FOR THE CURRENT ACCOUNT Our forecast that residential investments will decline supports our forecast that the current account will stabilise going forward. The sum of gross fixed investments (real domestic savings) and the current account (financial savings for the total economy) has increased over the last 3 years, even thought the ratio is slightly lower than before the financial crisis. Gross national savings in Sweden is among the highest among OECD countries even though it has declined slightly compared to (Norway and Switzerland have higher national savings than Sweden). CURRENT ACCOUNT CONCLUSIONS: The downward trend for the current account balance over recent years is a partly a result of a weak international recovery, but rising imports driven by strong domestic demand is the most important reason. The upturn in domestic demand is primarily explained by rising residential investments, which made fixed investments share of GDP rise to record levels in Total savings in Sweden has increased over the last 3-4 years as financial savings (current account) has been more than offset by higher fixed investments. The combination of stronger international growth and a decline in residential investments suggest that the current account balance will stabilise or even increase slightly in 2018 and

18 SEB SEK Positioning & Flow analysis International accounts behind SEK weakness FOREIGN ACCOUNTS SELLING WHILE DOMESTICS ARE EITHER BUYING OR NEUTRAL We have analyzed several indicators to identify how various key market participants are positioned, and their attitude towards SEK. Actions by Key Players in SEK vs EUR Domestic As a result, we conclude that: Both foreign fundamental and systematic investors have been selling SEK and are, compared to previously, excessively short SEK. Domestic corporates are naturally short SEK but have been buying SEK, especially since EUR/SEK rose too high in February. Domestic institutions sold SEK Q but have been fairly inactive/neutral in Foreign Systematic Investors Institutions SEK Domestic Corporates Foreign Fundamental Investors Selling SEK Buying SEK 18

19 SEB SEK Positioning & Flow analysis Domestic accounts may provide further positive SEK support CLIENT FLOWS INDICATE DOMESTIC PLAYERS COULD SUPPORT SEK Regarding weekly sentiment, most of the time Swedish companies have remained net buyers of SEK, especially since February when EUR/SEK once again reached long-term levels sufficiently attractive to sell foreign currency holdings and buy SEK. On the other side, institutions have more often been net sellers of SEK, although so far this year the flow size has been comparatively small (usually at or below average) and sentiment low CONCLUSION 1. Swedish companies tend to take advantage of weak SEK levels and may therefore be expected to continue to generate SEK positive flows around current high EUR/SEK levels 2. Since Swedish institutions have not participated much in driving EUR/SEK higher in 2018, position unwinding is unlikely to support SEK 3. On the other hand, seemingly Swedish institutions are skeptical regarding the late push higher in EUR/SEK, as they have not participated. Instead, they will probably stand ready to buy into a correction lower 19

20 SEB SEK Positioning & Flow analysis International speculative accounts are (very) short SEK POSITIONING MEASURES SHOW FOREIGN INVESTORS ARE VERY SHORT SEK Combining different measures for positioning indicates that, from an historical perspective, international speculative accounts hold an unusually large short SEK (long EUR) position According to our proprietary speculative SEK index, fundamental/discretionary investors hold a multiyear large short SEK position Option-based investors are also short SEK as indicated by the risk reversal which currently trades at a close to a 2 year high Systematic investors (e.g. CTAs) also hold an excessively short SEK position according to our trend gauge, which also stands at a high level CONCLUSION Clearly the international speculative market has been the driver for a weaker SEK. Our client survey also showed that domestic accounts remain long validating these findings in SEB Speculative SEK flows of continued foreign SEK selling. 20

21 SEB SEK Long-term valuation Relative ULC-approach to determine the equilibrium exchange rate Traditional theories behind the equilibrium exchange rate, such as Purchasing Power Parity (PPP), considered the relative price of goods and services. In its purest form this theory defines the equilibrium exchange rate as the rate at which a basket of goods (given the prevailing exchange rate) is equally priced in two different countries. However, in a globalized world the relative ULC, reflecting a country s competitiveness, is probably a more relevant measure of the long-term equilibrium exchange rate for countries that compete on the global markets in production of goods and services. ULC is simply a function of labour compensation and productivity growth. Over the longer term it is probably impossible for an exchange rate to deviate substantially from the level suggested by relative unit labour cost (ULC), as this would undermine a country s competitiveness. As a comprehensive measure of a country s competitiveness, relative ULC can be used to estimate the equilibrium exchange rate between two currencies. This is simply the nominal exchange rate that offsets changes in relative ULC so that competitiveness is maintained over time. Nonetheless, it appears it doesn t work for all currencies as there are examples of commodity currencies like the AUD and NOK, where favourable terms of trade due to surging commodity prices seems to have undermined the relationship between the countries competitiveness (ULC) and the exchange rate. High commodity prices make these countries afford weak competitiveness outside the commodity sector. 21

22 SEB alt. equilibrium exchange rate in EUR/SEK Relative ULC has moved EUR/SEK from 8.70 in 2006 to 9.70 Comparing ULC between Sweden and the Euro area since the middle of the 1990 s, ULC in Sweden has increased faster than in the Euro area and most of it took place in recent years. The reason is simply that, with the exception of 1998, labour compensation has increased more rapidly in Sweden than amongst Euro countries. Up until 2010, Swedish labour compensation rose at an annual rate of 3.9%. Subsequently, the rate has slowed to 2.6%. Labour compensation in Euro area countries has been largely stable or has decelerated since the start of the Euro crisis. Between 1996 and 2009, it increased at an annual average rate of 2.3%. However, since 2010 it has fallen to 1.6%. Partly, faster growth in Swedish labour compensation has been neutralized by more rapid Swedish productivity improvements, particularly between 2002 and 2008 when Swedish ULC fell compared to the Euro area despite faster growth in labour compensation. The chart (bottom right) shows the equilibrium exchange rate between Sweden and the Euro area based on the development in relative ULC between the two countries since Assumption: The actual exchange rate fluctuates around the equilibrium exchange rate => mean value or a linear trend over time is the same for the equilibrium exchange rate as for the actual exchange rate. Equilibrium exchange rate 9.70 in EUR/SEK 22

23 SEB alt. equilibrium exchange rate in USD/SEK Relative ULC in USD/SEK has been quite stable over time Comparing ULC between Sweden and the US since the middle of the 1990 s, ULC in Sweden has increased slightly faster than in the US but differences are very small. Although the nominal USD/SEK exchange rate appears to fluctuate much more over time than EUR/SEK, the equilibrium exchange rate seems fairly stable over time. The equilibrium exchange rate suggested by this relative ULC approach is currently According to this approach the equilibrium exchange rate reached lows at 7.17in 2006 after a period of relatively strong Swedish productivity growth between 2002 and However, during the years following the financial crisis ( ) relative productivity growth was much weaker in Sweden while Swedish labour compensation increased faster than in the US at the same time. As a result, the relative ULC in Sweden rose compared to the US. This caused the equilibrium exchange rate in USD/SEK to increase to 8.12 in 2013, where it has remained fairly stable as relative productivity growth and improvements in labour compensation have been largely the same in both countries. The result is very much the same when compared with the Euro area, with the equilibrium exchange rate hitting a USD/SEK low in 2006 at 7.18 after Swedish productivity outperformed. 23

24 SEB Long-term valuation (SEBEER) EUR/SEK ULC-approach differs significantly from SEBEER-model Previous long-term fair value estimates for EUR/SEK and USD/SEK were based on a combination of the long-term trend in the nominal exchange rate and the SEBEER-model The SEBEER-model is based on four explanatory factors, relative inflation, relative productivity, relative interest rates and relative terms of trade and is estimated on a panel of 28 currencies against the USD since 1980 The long-term fair value estimate in EUR/SEK according to this approach is 9.13 and 7.75 in USD/SEK While the USD/SEK fair value estimate is quite close to the rate suggested by relative ULC the EUR/SEK estimates deviates substantially. While there is no obvious way to determine in an objective way what approach that gives the correct equilibrium exchange rate we should perhaps consider the range

25 Event study: EUR/SEK and previous rate cycles EUR/SEK has on average declined by 5% 6mths before first Rix hike 2018: RELATIVE POLICY EXPECTATIONS SEEMS SEK SUPPORTIVE The Riksbank has in the past years strongly argued that its monetary policy cannot deviate too much from ECB s monetary policy as it would strengthen the krona too much and lower import prices and inflation. Since 2014, the Riksbank has lowered its repo rate by 85bps more than the ECB (deposit rate), while both central banks have bought large amounts of bonds. During this period EUR/SEK have increased by almost 20%. In this study, we analyse the behaviour of EUR/SEK in the 4 rate hike cycles since Although many factors will determine exchange rate movements, an overall analysis suggest that in all 4 cycles the krona has strengthened in the 6 months preceding the first hike. The magnitude of the appreciation and the path differ between the cycles, but in 3 out of 4 cycles, EUR/SEK seems to have reached a peak 6-7 months before the first hike and then declined on average by 5%. To better find a potential link between EUR/SEK and changes in monetary policy expectations, we analyse Riksbank s 4 hike cycles in detail by reading historical Monetary Policy Reports and combine it with market pricing according to future contracts that we use as a proxy to capture market expectations. According to relative policy expectations, EUR/SEK should actually have traded higher in recent years but based on our forecast that the Riksbank will hike around 9mths before the ECB, EUR/SEK is expected to fall in However, given that the pace of Riksbank s hikes will be much slower than in previous cycles, EUR/SEK should decline only gradually and trade close to our forecast at 9.80 at the end of this year. 25

26 Event study: EUR/SEK and previous rate cycles : Relative expectations matter for SEK The orange line in the upper chart shows the difference between the 4th SEK FRA and the corresponding Euribor futures contract, which we use as a proxy for relative monetary policy expectations in 1y. Relating these movements to EUR/SEK, the chart shows that the two series follow each other well during certain periods but also differ quite noticeably in other periods. The correlation is, however, fairly high (-0.51) and a regression with relative expectations as the explanatory variable has an R2 of almost 0.3. Analysing the cycles separately, the lower chart to the left shows relative policy expectations and EUR/SEK from 2008 to 2012, together with the actual difference in policy rates (green line). During the financial crisis the krona weakened substantially but started to appreciate in March Some of the weakness was probably exaggerated by crisis, but in general EUR/SEK seemed to have moved largely in line with changes in monetary policy expectations. After the Riksbank had lowered its repo rate by 125bps more than the ECB from end-2008 to mid-2009, the Swedish economy stabilised in 2009 and the market increased its expectations that the Riksbank would raise its rate before the ECB. In the year ahead the Riksbank s first hike in June 2010 relative expectations increased by 85bps and EUR/SEK declined by 12%, from around to In total, the Riksbank hiked the rate by 125bps from June 2010 to February 2011 while the ECB held its rate steady at 1%. During this time EUR/SEK continued to fall but stopped 2-3mths before the ECB s first hike (April 2011) as expectations also shifted. In general, EUR/SEK moved very close to policy expectations during this period and the correlation was above

27 Event study: EUR/SEK and previous rate cycles Mid-2000: Less focus on relative monetary policy Both ahead of and after the first rate hike in 2006, EUR/SEK fluctuated much less than relative expectations suggested. Despite the Riksbank lowering its repo rate by 125bps in while the ECB kept its policy rates unchanged, EUR/SEK hardly moved. Instead, it traded much stronger than might have been anticipated based on relative expectations. In 2006, the Riksbank began to hike largely in tandem with the ECB and EUR/SEK remained relatively stable. However, from mid-2007, relative expectations started to increase, but this time without producing larger movements in EUR/SEK, finally causing the series to converge again in mid One possible explanation why the krona did not weaken more in may be that both Swedish and international economic conditions during this period were relatively sound, and that the Riksbank s cuts attributed to surprisingly weak inflation which came relatively late in the cycle when markets were instead expecting higher rates in both Sweden and abroad. Although relative expectations had to adjust to actual rate reductions, the market may have continued to focus on upcoming hikes, resulting in a smaller impact on SEK. The hiking cycle that began in Mar 2002 included only two rate increases and followed a relatively short period with a lower rate after a 50bps cut in Sept 2001, which was closely connected with the terrorist attack in New York. Before the cut in September 2001, the Riksbank had been in a hike cycle and raised its repo rate by 135bps between November 1999 and mid As these two cycles are closely connected, we have decided to analyse them together. 27

28 Event study: EUR/SEK and previous rate cycles : SEK more sensitive when relative expectations rise Analysing the entire period from 1999 to 2004, relative expectations apparently captured reasonably well at least some of the movements in EUR/SEK. However, the decline in EUR/SEK that began in early 1999, after the ECB had cut its rate while the Riksbank left its own unchanged, was stronger than implied by expectations. Also, when the ECB started to hike more rapidly than the Riksbank in 2000, the krona did not depreciate as much as expected. This development is largely in line with the pattern seen in mid-2000, which may suggest that EUR/SEK is less sensitive when ECB hikes faster than Riksbank. After the terrorist attack in September 2001, the Riksbank together with many other central banks reduced its rates. During the fall, the ECB continued to lower its rate, while the Riksbank kept its own unchanged which caused relative expectations to increase and from September 2001 to March 2002, EUR/SEK depreciated by 10%. After the first hike in March 2002, the Riksbank hiked one more time but as relative expectations stabilised, EUR/SEK moved relatively little. While it remains difficult to draw any firm conclusions from analysing previous hiking cycles in detail (especially as we only have 3-4 cycles to consider), various patterns still seem to exist. Over time relative policy expectations and the development in EUR/SEK apparently converge. Indeed, this connection looks even stronger when markets expect the Riksbank to increase its rate faster than the ECB, rather than the converse. Analysing the four occasions on which the Riksbank has hiked its repo rate, alterations in relative expectations nine months before the first hike also appear a likely explanation for much of the change in EUR/SEK during the same period (lower chart). 28

29 Event study: EUR/SEK and previous rate cycles Krona to strengthen only gradually 2018 based on our CB forecasts PREVIOUS HIKE CYCLES SUGGEST EUR/SEK TO TRADE AROUND 9.80 DEC-2018 Relating this analysis to movements in recent years, the large decline in relative expectations that followed the Riksbank s shift in policy in 2014 would, in line with previous cycles, actually suggest a weaker krona. Since 2016, expectations that the Riksbank will hike before the ECB have begun to increase. However, they have still not caused a higher EUR/SEK and the two series have again converged. In the past few years, changes in market expectations and EUR/SEK are reminiscent of movements seen in , when the Riksbank lowered its repo rate when growth was strong. In recent years, both we and financial markets have expected that the strong Swedish economy has justified a shift by the Riksbank. However, as the country s central bank has continued to focus only on inflation, we and the market have had to adjust our expectations over time. The fact that most market participants probably did not expect the Riksbank to go as far as it has, continuing its prolonged expansive monetary policy, is probably a reason why the krona did not depreciate more in Looking ahead, the situation now reminds us more of the hiking cycle in We expect the Riksbank to hike in December this year, 6-9 months before the ECB. If we are correct, relative expectations should continue to increase slightly. However, compared to previous cycles, the Swedish central bank is expected to hike more slowly (1 hike in 2018 and 2 hikes in 2019). By the end of 2019, the rate differential vs. the ECB is forecast to have increased by only 50bps. Relating expected relative changes to previous cycles, data suggest EUR/SEK should decline by 4-5% over the next nine months. Based on today s value, this implies that EUR/SEK should trade around 9.80 by the end of this year. 29

30 Recommendations & FX forecasts Trades & Recommendations We expect the trade-weighted krona to appreciate cautiously in the coming 6-9 months We have initiated a short EUR/SEK trade recommendation in FX Ringside (3 April) at with a target of 9.80 and stop loss (daily close) above SEB targets EUR/USD at 1.30 in 12 months time and a lower EUR/SEK -> USD/SEK 12 months forecast is found at 7.45 (current 12 mth fwd rate at 8.05). Hence we recommend to sell USD/SEK or to hedge USD exposures (stop above 8.60). We would look to buy double-no-touches with a 3-9 mth duration in NOK/SEK and GBP/SEK The main risk as regards our now bullish SEK outlook remains the long-term subdued outlook for Swedish inflation and a dovish Riksbank unable to deviate from soft ECB policy

31 Contacts Carl Hammer (Editor) Richard Falkenhäll Lina Fransson Håkan Frisén Olle Holmgren Karl Steiner

32 This report is produced by SEB for institutional investors only. All information has been compiled in good faith from sources believed to be reliable. However, no representation is made as to its accuracy or completeness. No liability is accepted for any direct or consequential loss resulting from reliance on this document. Changes may be made to opinions or information contained herein without notice. The report is the intellectual property of the bank. Quoting from, referring to, altering or editing in parts or in its entirety of the report is not allowed without the prior written consent of the bank. This material does not constitute investment advice and is being provided to you without regard to your investment objectives or circumstances. SEB, its affiliates or employees may buy or sell, hold long or short positions in, or act as principal in, the financial instruments referred to herein or may from time to time perform or seek to perform investment banking or other services to the companies mentioned herein. For more information see 32

33 Blank page 33

34 Sweden: Election 2018 A close race according to the polls EVEN RACE ACCORDING TO POLLS According to polls, the race will be an even one with no majority for the traditional blocks (Social democratic led left-green alliance and the centre-right Alliance that ruled between ). 2 parties are currently polling below the 4% threshold to get seats in parliament. It is unusual though for smaller parties to actually be kicked out of parliament (only happened 3 times since 1945 and 2 of those has made a comeback). Only thing that have some degree of certainty is that neither block will get an own majority. It will, again, be difficult to form a strong government after the election. LACK OF STRONG GOVERNMENT, STRUCTURAL REFORM IN WAITING MODE After 8 years of minority governments, the need for structural reform and agreements over the traditional political blocks become more and more acute. We have seen broad policy agreements during the last years in less politically sensitive areas such as defence, energy and immigration. In more politically sensitive areas such as labour market policy, taxes and housing, broad agreements have been difficult to reach As the pre-election atmosphere is expected to be characterised by confrontation and tactical positioning, we expect no near term guidance on how parties will handle structural issues ahead. Even race according to polls Poll of polls (Sv. väljaropinion) March -18 Elec. -14 Social democrats Environment party Left party S+G+L Moderates Centre party Liberals Christian democrats M+C+L+CD Sweden democrats