In our updated forecasts for 2010-11, we continue to assume that lawmakers will pull back from the abyss of a policy-induced downturn in the economy and temporarily extend the bulk of the tax cuts. Alternatively, the Administration is considering a variety of ‘targeted' measures to boost job growth and help the economy. However, officials are avoiding using the word ‘stimulus' to describe any such measures, reflecting the public perception that previous action did little to help the economy. Moreover, any proposals this close to the mid-term elections will undoubtedly face tough sledding in Congress. Either way, if gridlock is not resolved soon, and taxes on all income groups rise, virtual fiscal drag will become reality. That will potentially trim about 0.75pp from growth in 2011.
Bimodal backdrop. This policy uncertainty is intensifying against the broader backdrop of a slowing economy, with the risk that gridlock could stymie an effective response. Indeed, we now see 2H10 real GDP growth at 2-2.5%, implying upside risks for unemployment and downside risks for inflation. Barring a resolution to the tax debate, that puts the burden for supporting growth and reducing risks of still-lower inflation on the Fed. Fed Chairman Bernanke last week clearly indicated a bias towards additional monetary ease, most likely in the form of further asset purchases.
Bimodal - and asymmetric - outlook for yields. Thus, for bond investors we see a bimodal outlook, but one that is asymmetric. Policy choices over the next few months will dictate the outlook for longer-term yields over the next year or so, but the longer-term direction is higher. Policy gridlock and additional economic weakness will put more pressure on the Fed to purchase securities; such purchases may push 10-year Treasury yields to 2%. In that case, the outlook for yields will mostly depend on how low officials think they must go to fulfill their dual mandate. But breaking the policy logjam in Washington may revive risk appetite and the economy. In that case, recovery and a bottoming in inflation will restore market forces as the dominant drivers for the back end of the yield curve, and push 10-year yields past 4% late in 2011.
Has fiscal drag already slowed the economy? We think the effect has been modest. Judging by data through 2Q and estimates of the impact of last year's stimulus package under the American Recovery and Reinvestment Act (ARRA), the primary source of federal fiscal drag came this spring from the expiration of extended unemployment benefits and aid to state and local governments; those transfers have since been reinstated. Indeed, obligated (committed) ARRA outlays through June 30 amounted to $403.8 billion (of the $499 billion total), while outlays amounted to just $257.3 billion. So there is more spending, especially for infrastructure, in the pipeline, and fiscal thrust has only just peaked. To be sure, according to CBO estimates, the fiscal impact of ARRA on growth has turned slightly negative in 2H (CBO's low estimate suggests that the change in impact will be -0.6% of GDP between 2Q and 4Q10). But in our view, the ‘multipliers' and lags between thrust and impact derived from models employed to make those judgments aren't constant; the multipliers have increased (and the lags may have shortened) as financial conditions improved. If we are correct, a switch from lower to higher multipliers would reverse the sign of that impact.
So why the slowdown? In our view, the main culprit for weakness relative to our forecast is less-than-expected support from global growth, which we had expected to offset a slower pace of domestic demand. In addition, the reversal of the first-time homebuyer tax credit boosted 2Q growth by about half a point at the expense of 3Q. Net exports have been a drag on output this year, and even a positive swing in 2H won't make up for the 2Q downdraft, when surging imports substituted for US output. On the export side, global domestic demand growth among our trading partners seems to be weaker than we had expected.
The coming trade tailwind. Incoming data reinforce the view that foreign trade has been a key culprit behind the slowdown but that this factor could start to swing in the other direction - albeit in a small way. Exports slowed in June and the diffusion index of export orders in the August ISM, while still at a strong 55.5%, was the lowest in 2010. The June surge in imports likely reversed in the past couple of months, so that we expect net exports to contribute about 0.5pp to growth in 2H10. Indeed, two factors - one on each side of the ledger - should turn the impact of trade from strong headwind to breezy tailwind: Still-hearty gains in global domestic demand should promote vigorous growth in exports, while a sharp moderation in US demand will slow imports dramatically. In particular, we think that four global factors will boost US exports: a global capex upswing, an uptrend in infrastructure building, especially in EM economies, the increasing influence of the EM consumer, and the Russian drought and wheat export ban that will boost agricultural exports. Even so, the expected half-point contribution will unwind only a small fraction of the 3.4% annualized drag in 2Q.
Two-track economy. In contrast, sluggish vehicle sales, construction outlays and personal income evince a two-track economy - domestic demand is growing slowly, so the economy is still dependent on growth overseas to lift it above trend. Job and income growth are clearly needed for a self-sustaining recovery, and recent data barely qualify: Private payrolls gained an average 78,000 in the three months ended in August, and thanks to a rising workweek and improving hourly earnings, ‘core' after-tax wage and salary income is now running at a 2-2.5% pace. Nonetheless, leading employment indicators - a rise in the factory workweek and overtime, a rebound in temporary help jobs, and the 369,000 rebound in employment measured by the household survey (adjusted for population and consistent with payroll concepts) - are encouraging signs that more improvement is coming.
Financial conditions improving... Partly offsetting the headwinds of policy gridlock, financial conditions are slowly improving. Willingness to lend to consumers has improved, judging by the 16-year high in that metric in the August Fed Senior Loan Officer Survey. Lending standards - even to small businesses - are beginning to ease. Mortgage rates have declined significantly to a record low 4.3% in early September for 30-year, fixed-rate loans. As originators hire staff to process new loan applications for the current refinancing wave, the spread between primary and secondary rates may compress further. That refi wave will further reduce household debt service, making many consumers more creditworthy and creating a virtuous circle for credit availability. The looming supply of investment grade and high yield corporate debt slated for September, estimated as high as $100 billion, testifies to ultra-attractive financing terms for Corporate America. All these indicators point to support for credit-sensitive demands.
....but not all the improvement is reaching borrowers. Nonetheless, some critical institutional hurdles are limiting the benefits to mortgage borrowers of lower rates and an easing in credit conditions. Stringent criteria for refinancing and capacity bottlenecks have limited refinancing opportunities for many borrowers and kept effective mortgage rates well above secondary market yields. Originators continue to be hamstrung by ‘putback' risk - the ability of the GSEs to force the originator to repurchase loans that become delinquent. Industry experts believe that more aggressive putbacks have led to tighter underwriting and have spurred counterparties further down the mortgage food chain to toughen standards on one another. Of course, all of this is part of the more general disruptions to refinancing that we have been highlighting of late. Indeed, our mortgage strategists note that the primary-secondary spread hit a new cycle-high near 125bp this week, up from the 80bp gap seen in the spring. 30-year mortgage rates have lagged behind the plunge in MBS yields to record lows in recent months.
Below-trend growth now, moving back to or above trend. In our view, four factors will promote a return to stronger growth in 2011: 1) the debate over extending the expiring tax cuts will be resolved favorably by early next year; 2) coming stability in home prices and stronger household balance sheets should prompt consumers to spend a bit more; 3) continued low rates near term should facilitate mortgage refinancing and reduce housing risks; and 4) a return to stronger growth in Asia will sustain growth in US exports.
Tepid growth in the near term implies a higher unemployment rate at year-end, to 9.7% rather than the 9.4% we had assumed previously. Other measures of slack may either stall or perhaps widen as well. For example, with housing demand weaker than expected, there is a risk that vacancy rates will rise again. The implications for inflation are important: While we still think core inflation has bottomed, more slack will mean that ‘speed' effects - the effect of shrinking slack on pricing and inflation - will fade, and the rise from today's 0.9% CPI likely will be even slower, keeping inflation below the Fed's comfort zone for longer. Measured by the PCE price index, the Fed's preferred gauge, we now see core inflation back in a 1.5-2% range by 2H11, compared with our previous forecast of 2%.
Fed biased to ease, but action unlikely in September. Given policy uncertainty, and with below-trend growth threatening a higher unemployment rate and lower inflation, Fed officials are weighing additional anti-deflation insurance. Friday's better-than-expected employment report makes it unlikely that the Fed will implement new monetary stimulus measures at the September FOMC meeting. But such action remains a possibility at some point during the next several months if the incoming data suggest that the underlying pace of growth is slipping below the economy's escape velocity of +2% or so.
In summary, we see a bimodal outlook. We expect political gridlock to be a drag on growth and inflation in the near term. It's not our base case, but full sunset of all tax cuts could pare 0.75pp from 2011 GDP growth. If that scenario looks likely, yields may drop to 2% as the Fed resumes asset purchases. In our base case, growth should rebound in 2011, especially if policy action is favorable, and the ultimate direction for yields is higher - to 4% or more by late in 2011.
The Treasury yield curve over the past week significantly extended the big steepening move seen in response to Fed Chairman Bernanke effectively announcing at Jackson Hole that the Fed has an easing bias, as the first big positive surprise in a key economic release in a good while in the manufacturing ISM report and a better-than-expected, though certainly still sluggish, employment report further eased pessimism about the economic outlook. Overall, the past week's economic news was more upbeat but still somewhat mixed, though this marked a big shift at least from the run of dismal economic news over the prior couple of weeks that added to the sour impression investors largely had of the August 10 FOMC statement. The medium-term growth outlook still looks sluggish, but strength in the manufacturing sector globally points to robust trade growth, enough jobs and income are still being created to support a continued modest expansion in consumer spending - and early indications for the retail sales report in company reports on motor vehicle and chain store sales were positive overall - and incoming data the past week directly impacting our near-term GDP estimates left our 3Q forecast near +2%, disappointing but not the double-dip recession outlook that investors were increasingly starting to fear in August. The 2H growth trajectory of 2-2.5% we now project, down from the 3-3.5% we were looking for a month ago, would be slightly below trend, raising the risk of renewed disinflation from already uncomfortably low core inflation rates. And prospects for a pick-up back to above-trend growth over 3% in 2011 increasingly seem to hinge on the outlook for taxes next year, uncertainty over which is only rising as neither side has shown any inclination to compromise two months ahead of the elections. So, the prospect of renewed Fed easing in the form of ramped up quantitative easing certainly still remains a medium-term possibility. But the more upbeat initial run of key economic data for August has likely taken this off the table for the September 21 FOMC meeting.
On the week, modest short-end gains and long-end losses sent 2s-30s another 13bp steeper for a 26bp cumulative move in the six trading sessions since Fed Chairman Bernanke's Jackson Hole speech. The 2-year yield fell 5bp to 0.51%, 3-year 4bp to 0.79%, and 5-year 1bp to 1.48%, while the 7-year yield rose 4bp to 2.13%, 10-year 6bp to 2.70%, and 30-year 8bp to 3.78%. TIPS at the shorter end lagged, but longer-end issues also extended the outperformance seen after Bernanke's speech (implying a larger uptick in 5-year/5-year forward inflation expectations) as deflation fears continued to ease, with upside supported by a solid 10-year TIPS auction. The 5-year TIPS yield rose 2bp to 0.16%, the 10-year yield was steady at 1.02%, and the 30-year yield rose 5bp 1.70%. This left the benchmark 10-year inflation breakeven at 1.68%, up 6bp on the week and 16bp from the low close of the year of 1.52% hit August 24 after a two-week plunge of 33bp after the FOMC meeting. Lower coupon mortgages traded a bit better than Treasuries on the week, with current coupon yields ending little changed near 3.45% after hitting more record lows near 3.2% early in the week. While MBS yields have been trading near these record lows since mid-July, mortgage rates being offered to consumers are still only very gradually catching up as the heavily concentrated and shrunken post-crisis mortgage origination industry works through capacity constraints. Indeed, our mortgage strategists note that the primary/secondary mortgage spread actually hit a new high this week, having risen about 45bp since the spring lows to a well-above-normal level. Still, the persistence of the MBS strength is at least gradually dragging mortgage rates down, with Freddie Mac's national survey showing the national average 30-year rate at 4.32% this week, down a quarter-point from mid-July. And this is starting to drive a more pronounced mortgage refi wave that will add to the already well-advanced consumer balance sheet repair.
Risk markets responded very positively to the better-than-expected key August data results, especially the very surprising gain in the manufacturing ISM, to the extent that the miserable results of August were fully reversed in the first three days of trading in September. At the time of the futures close Friday, the S&P 500 was up 3.4% on the week, with a 5% gain Wednesday to Friday reversing the 5% drop posted in August. Financials were the best-performing sector, followed by consumer cyclical stocks and industrials, which were boosted by the better-than-expected chain store sales results and ISM. In late trading, the investment grade CDX index was 7bp better on the week at 104bp, also fully reversing August's losses with a 10bp tightening from Tuesday's close. The high yield CDX index tightened about 50bp on the week to near 550bp, also reversing last month's losses.
Economic data released the past week were mixed overall, which was a lot better than the previous recent run of consistently dismal results but still broadly consistent with our downgraded outlook for slightly below-trend GDP growth of 2-2.5% in 2H. There was the first major positive surprise in a key release in quite a while in the manufacturing ISM, an employment report that was better than expected but still quite sluggish in an absolute sense, decent early indications for retail sales in flat motor vehicle sales but better-than-expected chain store sales, a very weak construction spending report and a better-than-expected factory orders report that were offsetting and left our 3Q GDP forecast at +2%, and a soft non-manufacturing ISM report.
Non-farm payrolls fell a smaller-than-expected 54,000 in August after significantly smaller revised declines in July (-54,000 versus -131,000) and June (-175,000 versus -221,000). Temporary census jobs fell 114,000 in August, leaving about 82,000 left to be laid off in coming months, while private payrolls rose 67,000 on top of a 66,000 upward revision that brought average gains the past three months to +78,000, a bit better than the +50,000 average expected coming into this report. Upside was led by construction - +19,000 overall, with heavy construction and civil engineering up 11,000 and non-residential contractors up 18,000, while residential was down - as government infrastructure spending still seems to be picking up significantly despite the recent weakness seen in the construction spending report. Temp jobs (+17,000) were also up solidly and recession-proof healthcare (+40,000) posted a typically big gain. The weakest area was manufacturing (-27,000), but this was almost all in motor vehicles after a big gain in July, as there was much less seasonal downtime this year than normal, resulting in fewer temporarily laid-off workers needing to be rehired in August. Other details of the report were mixed. The unemployment rate rose a tenth to 9.6%, though this resulted from a surge in the labor force, as the volatile household measure of employment actually gained 290,000. The average workweek was flat at 34.2 hours, and total hours worked were flat. But manufacturing hours were up 0.1% even as a sizeable pullback in the auto sector reversed some of July's surge. Based on this upside, we see industrial production gaining 0.5% in July and manufacturing ex motor vehicle production surging 0.8%. Average hourly earnings jumped 0.3%, the biggest gain since January, so even with flat hours worked, aggregate weekly payrolls, a gauge of total wage and salary income, rose a decent 0.3%.
The strength in August manufacturing output implied by the employment report was also seen in a much stronger-than-expected result for the manufacturing ISM. This followed a better-than-expected Chinese PMI and German Ifo, indicating that global manufacturing activity and thus most likely global trade remained quite robust through the summer. This should help to provide some external support to 2H US growth as domestic demand slows - a divergence highlighted by the much more sluggish non-manufacturing ISM result for August. The composite manufacturing ISM index rose 0.8 points in August to a robust 56.3, far stronger than suggested by uniform softening in the regional surveys. Underlying details were also strong, with upside in the composite driven by a gain in production (59.9 versus 57.0) and a rise in employment (60.4 versus 58.6) to a level that hasn't been exceeded since 1973. The orders index (53.1 versus 53.5) was little changed at a modestly positive level. Expansion broadened by industry, with 11 of 18 sectors reporting growth in August compared with 10 in July when auto sector output was particularly strong. And anecdotal comments highlighted in the report were upbeat. In contrast, the composite non-manufacturing ISM index fell 3 points in August to 51.5, a low since January, as the employment index (48.2 versus 50.9) fell back into negative territory after a slightly positive reading last month, and growth in business activity (54.4 versus 57.4) and orders (52.4 versus 56.7) moderated. In line with the muted overall index, 9 of 18 sectors reported growth in August and 8 contraction after 13 sectors grew in July, and comments from survey respondents highlighted in the report were somewhat cautious.
Although the early August numbers were somewhat encouraging overall, they were consistent with a continued muted recovery for the overall economy even as the far-outperforming manufacturing sector is moving into the second year of a more typical V-shaped rebound. And mixed data released bearing directly on GDP left our estimates little changed. The weak July construction spending report pointed to a bigger 29% plunge in 3Q residential investment after the 2Q surge and a flattening out in state and local government construction spending after a strong 2Q. But the personal income report for July and the decent early indications for August boosted our consumption forecast marginally to +2.0%, the factory orders report pointed to higher growth in 3Q equipment investment near +6.5% and a slightly bigger 0.6pp boost to 3Q from inventories, and the much-smaller-revised declines in state and local government jobs in the employment report offset the weaker construction outlook and left our overall forecast for state and local government spending in 3Q near -0.5%. Taken together, we now see 3Q GDP running at a fractionally higher +2.1% instead of +2.0%, while we look for 2Q to be revised down another tenth to +1.5%.
The economic calendar in the upcoming week is unusually quiet after the Labor Day holiday Monday, which will leave focus in the Treasury market on supply through much of the week, 3-year, 10-year and 30-year auctions Tuesday, Wednesday and Thursday. The only data release of much note is international trade for July. This is a very important report this month, since it could have big implications for 3Q GDP. After the near-record 3.4pp drag on 2Q GDP growth as real imports soared 32%, in 2H, we are looking for slowing domestic demand to sharply reduce import growth and for export growth to remain solid, resulting in a half-point add to GDP. To get there arithmetically in 3Q from a very negative starting point at the end of 2Q, we're forecasting a big partial reversal of the spike in the June trade deficit in July. We look for the trade deficit to narrow $4 billion in July to $46 billion, reversing half of the $8 billion surge in the deficit in June, with exports expected to rebound 2.0% after falling 1.3% and imports to dip 0.5% after surging 3.0%. On the export side, factory shipments figures and industry data point to a rebound in capital goods, led by high-tech products and aircraft, and we expect agricultural products to also rebound after a weak result last month. On the import side, Energy Department figures point to upside in petroleum products, but this should be offset by partial pullbacks in consumer goods and autos after enormous gains last month.
We visited Warsaw on September 2 and had meetings with local economists, the NBP, a political analyst and the Ministry of Finance. Please find below our key takeaways. Feel free to call for more colour.
Growth outlook - still in a happy place: The 2Q data continue to show Poland growing at a healthy pace, with headline growth at 3.5%Y, higher than our and consensus expectations of 3.2%Y. In sequential terms, growth of 1.1%Q (sa) after 0.7% in 1Q suggests that momentum picked up in 2Q. Household consumption continued to grow nicely, up 3.0%Y (+0.7%Q); investment also rebounded strongly (+8.6%Y, or +12.9%Q) after a weak 1Q, and exports and imports were both up a lot, also due to a base effect (+17.0%Y and 18.2%Y, respectively).
In terms of contributions, domestic demand is back firmly in the driving seat, adding 3.8pp to overall growth. Of this, 50% came from household consumption, 10-15% from government and the rest from investment (mostly inventories). Interestingly, and likely in contrast with the Czech Republic and Hungary, the contribution from net exports was negative (-0.3pp), for the first time since the downturn began. This is yet another sign that consumption and imports are normalising and the economy is returning to trend growth. True, with credit flows impaired, we are unlikely to see 6-7% GDP growth again any time soon. However, Poland stands out to be in pretty good shape, with very limited slack, if any (it never entered recession), and more insulated from any euro area slowdown: its external sector is far smaller than in the Czech Republic or Hungary, with exports/GDP at 40%, versus 80% in Hungary and 70% in the Czech Republic. Perhaps less encouraging, it is possible that such a powerful inventory boost may not be sustainable. But overall, growth of 3.2% or higher in 2010 as a whole looks well within reach. For 2011, locals still expect that strong infrastructure investment related to the Euro 2012 championship will boost growth. This is also the reason why the central bank's latest forecast sees a ‘hump' in GDP growth over the next two years, with the economy slowing in 2012. Risks to our current 3.4% forecast for GDP growth for 2011 look skewed to the upside.
Monetary policy - hawks in charge, for now: There is clearly a great deal of uncertainty around the new MPC and its reaction function. Given that the whole Council is new, the exact hawks/doves split is not entirely clear and local analysts are still trying to figure out individual members' profiles. That said, the overall sense from our recent meetings in Warsaw is that our long-standing call that the rate-hiking cycle will start later this year is still on track. If anything, the tone was overall more hawkish than we expected. The NBP does not feel tied down to the outlook for ECB rates, and can start its own normalisation process well ahead of the ECB. In terms of communication, the elimination of the policy bias from the statement which took place in June was probably no accident. Some senior members on the board feel rather strongly that communicating too openly with markets and providing too many cues as to upcoming changes in policy removes a ‘surprise factor' which is key in affecting expectations.
While a start of normalisation in interest rates appears likely, an aggressive tightening cycle seems unlikely to us, as inflation pressures from the labour market are very different than they were in 2007-08. Polish firms have hoarded labour to a much greater degree than in the previous recession, so we are likely to see a slower employment recovery also. Further, labour shortages of the type we saw in 2007-08 (also due to large migration flows) seem less likely. Our current rate forecast of 100bp of interest rate hikes by end-2011 looks broadly fine, with perhaps some upside risks for next year.
In terms of fine-tuning our near-term policy rate call, we would not rule out a 50bp rate hike to start the cycle, followed by some months of rates on hold. Our central case, however, is for a 25bp hike in October, though September is also a possibility. This week, the MPC will meet the previous members of the Council for an exchange of views. That this is happening at this juncture may well suggest we are at a turning point in policy. We also note a sudden shift to a more hawkish tone by several Council members: Gilowska said that policy should not remain "static" for too long, Kazmierczak said that the MPC may have to discuss rate hikes soon, Bratkowski argued for a hike of up to 75bp to keep inflation expectations in check, and Zielinska-Glebocka revealed that the MPC already voted on a rate hike in August, which was a surprise. More hawkish monetary policy reinforces our bullish view on PLN, which we currently see at 3.70 versus EUR by end-2011.
Fiscal policy - do not expect fireworks: The overall philosophy of the government continues to be to implement as little tightening as possible in order not to upset voters ahead of two important elections: the local elections later this year (November/December) and the parliamentary elections which will likely take place in 4Q11. It follows that real fiscal tightening will not take place, and the government will instead continue on its baby-steps policy approach: the 1% VAT hike, freeze in public sector wage bill and cap on discretionary spending (25% of total) to CPI+1% are hardly draconian, especially in light of what other governments in CEE are implementing. The contingency plan seems to be to increase VAT further if the budget deficit is wider than expected. The bank tax, which has recently been mentioned in the press, is still in its early stages and we are not sure it will ever go through. Of course, more will be revealed in October, with the 2011 budget draft. But no one should expect fireworks, we think. The reformist wing of the PO seems frustrated with such slow progress, but the PM remains focused on the real prize: winning the 2011 elections by a large margin. Unless ratings agencies and markets suddenly end their love affair with Poland, there is little incentive to consolidate the debt. We continue to think, however, that the authorities are simply delaying the inevitable. The debt dynamics continue to show a steady rise in debt/GDP, which we think will ultimately trigger draconian cuts and in the extreme (debt/GDP above 60%) forbid any borrowing whatsoever. To ensure that the constitutional debt thresholds are not breached, the authorities only have two choices, in our view: i) tighten policy starting in 2012, meanwhile using one-off measures and privatisation to avoid a breach of the 55% threshold; and ii) change the domestic debt definition with a simple majority vote to a lower number that takes into account pension reform costs. The latter option would provide breathing room, but would be a last resort, we think, due to the potential damage to credibility (see below).
Pension costs and EU lobbying: During our visit to Warsaw, we discussed at length the issue of pension reform costs. On a like-for-like basis, the Polish authorities argue that the deficit to GDP ratio should be around 1.5% lower, and debt/GDP should be around 40%, rather than 54%. The reason is the following: in Poland, new entrants into the pension system contribute into a first pillar (ZUS, state-managed) and a second pillar (OFE), which is made up of private pension funds (we ignore a voluntary third pillar in this discussion, as it is small), which invests mostly in Polish bonds (60% of total) and equities (30%). ZUS de facto works as a Pay-As-You-Go system, where current taxpayers save for their retirement, but these savings are used to pay for current pensions. The OFE portion of the system is made up of individual pension accounts. These private, regulated pension funds invest this money to (mostly) purchase government bonds. In countries which do not have a ‘second pillar', the state can count on higher contributions under the first, state-managed pillar (the equivalent of ZUS). Therefore, the central government has to issue less debt, although of course this does not make its system more sustainable: it is as if, instead of public debt, it issued IOUs to current taxpayers, or a ‘promise to pay'.
The Polish authorities believe that the second pillar introduces welcome competition and diversification, and they ought not to be punished (i.e., show a higher official public debt/GDP ratio) for reforming their pension system and introducing compulsory private saving schemes. We think that the Polish authorities have a point: their debt numbers incorporate officially at least some of the costs of ageing, whereas in other countries (most of Western Europe) this is not the case. Whether they can successfully lobby Eurostat to publish a supplementary table with ‘like-for-like' treatment of pension-related debt is a different matter. Ultimately, the Polish authorities could, if push came to shove, redefine the national definition of public debt (relevant in the fiscal rules) to account for the above. This would push them further below some crucial debt levels (55%, 60% of GDP) and give them more time to adjust. However, this would be a last resort, as the credibility costs associated with this move could be high.
Turkish citizens will be heading for a referendum on September 12 to vote on the future of an extensive package comprising significant changes to the constitution, which has been in place since 1982. The current constitution was put together two years after a military intervention in Turkey and it had long been criticised for a bias towards military influence on politics, weak protection of personal freedom and articles that could hinder the process of accession into the EU.
Why worry? While the referendum, in our view, is important in terms of how it could shape Turkey's future in the medium term, it might also have significant implications for the political environment in the short term, especially heading into the general elections scheduled for July 2011. That is why we wanted to focus on this again, at the expense of repeating some recently published views (see "Looking Back, Moving Forward", CEEMEA Macro Monitor, August 23, 2010). In our view, the market, especially non-resident investors, has broadly been ignoring the issue and assuming a comfortable margin of support at the referendum. This might well be the case, and we would assign a rather high probability to this; however, we believe that the outcome could have a significant influence on policy implementation. In fact, we believe that in the absence of clear support for the amendment package at the referendum (i.e., more than 60%), fiscal risks will rise noticeably amid escalation of political noise, not to mention the dilution of the government's focus from economic policy-making towards domestic politics.
Essentially, on September 12, the public will be voting for or against a set of amendments in the constitution that the governing party, AKP, has proposed. The amendment includes various changes that would bring the Turkish constitution closer to EU standards in terms of democratic principals and individual rights; it broadly removes the influence of the military on day-to-day politics. While this seems like a straightforward matter, there are certain amendments regarding the judiciary that the opposition is against. Some of the changes proposed by the draft amendment package call for a significant change in the composition of the high courts as well as the appointment process for the future members - and it is this which has drawn criticism from the opposition. In essence, the proposed constitutional changes would allow both government and parliament to be involved in the appointment of a certain number of judges to the Constitutional Court as well as the High Commission of Judges and Prosecutors - two bodies of the judiciary that had been perceived as key posts protecting the secular nature of the country.
In the early stages of the parliamentary voting process regarding the constitutional amendment package, one of the key changes in the draft package was rejected, and this was a first defeat for PM Tayyip Erdogan in terms of the proposed constitutional reforms. Essentially, the article in question was proposing to make political party closures, which had resulted in some 26 closures in the past, very difficult. Over the past two decades, these party closures resulted from court decisions based on verdicts of separatism and/or defending anti-secular practices that were not compliant with the constitution. In early July, a second blow to the constitutional amendment motion came from the Constitutional Court, which annulled some aspects of constitutional changes that were designed to curb the powers of judges and prosecutors. At this juncture, the referendum is now being seen as a race between the ruling party, AKP, and the opposition.
Scenarios: In our view, the outcome of the referendum on September 12 is seen as a reference event that might give significant shape to the political climate, and hence affect the fiscal picture, heading into general elections scheduled for July 2011. To us, as well as most political analysts, there are four main scenarios that are worth examining:
1) Strong support for a change (>60%): If the referendum attracts strong support, the market is likely to take that as a cue for continued support for the governing party and to assume an AKP victory at the general elections. That would mean a single-party government which might suggest no accelerated election-related fiscal spending. From the markets' point of view that would be good news and in our view the current pricing of Turkish assets is almost fully reflecting this. However, based on the publicly available poll results, we believe that such a comfortable margin of support is unlikely. For instance, the most recent poll by a well-regarded research company (A&G) predicts 45.2% support versus 44% rejection but with 10.8% undecided. Another poll by SONAR gave 50.9% to the rejection camp with 49.1% predicting support. Other polls that have appeared in the media have predicted varying degrees of pro- and anti- votes such as Andy-AR giving 53.4% in favour of the change and MetroPoll assigning 49.6% to the same camp. We are clearly not taking any of these poll results at face value and acknowledge the margin or error as well as the ample time for either of the camps to influence the results. However, we feel comfortable in predicting that the outcome is not a ‘done deal' at all.
2) A narrow support (slightly above 50%) could result in a more or less market-neutral impact, with AKP capitalising on the success, albeit still having doubts regarding a sure-fire victory in the general elections. Hence, the market might stay sidelined but the fiscal risks would still remain in the background. In fact, we tend to believe that despite the seemingly market-friendly outcome in the short term, a narrow victory on the part of AKP could cause heightened consciousness ahead of elections. Granted, a portion of the votes supporting the constitutional amendment proposed in the referendum will also be coming from the non-AKP camp (i.e., people believing in the necessity of a change in the constitution but not necessarily intending to vote for AKP in the elections). This could easily encourage the government to increase fiscal spending (helping to lower unemployment on a temporary but costly basis) and to delay politically unpopular reforms. With a quite strong track record on the fiscal side during 2002-07, the government, in our view, deserves the benefit of the doubt. However, the fiscal slippage witnessed during the general elections in 2007 and the local elections in 2009 suggests caution. More importantly, the recent (and last-minute) decision by the government to delay the implementation of the long-anticipated fiscal rule was a major disappointment to us. Based on the poll results at hand, we would assign the highest probability to this scenario at this stage. In line with this, we are ready to revise up our fiscal deficit to GDP forecasts both for 2010 (currently at 4.3%) and 2011 (currently at 4%), but will do so depending on the outcome of the referendum and its aftermath.
3) A narrow loss (slightly below 50%) would, in our view, indicate success on the part of the opposition and rising support for CHP (the main opposition party). In our view, the immediate impact of this would be market-unfriendly as it would raise fiscal concerns and bring post-election coalition scenarios to the foreground. In our view, this scenario is not priced at all and the realisation of such an outcome will most likely result in a significant market weakness.
4) Strong opposition at the referendum (i.e., rejection with more than 60%) would clearly be market-unfriendly as it would surely be perceived as a sign of a lack of support for the government, raising the probability of a coalition government following the 2011 elections. The other assumption attached to this outcome would be to expect the government to potentially accelerate spending and resort to populist policies to counteract loss of power. We do not have any data to justify such an extreme case outcome and we would assign a very low probability to it.
Strategy Implications
Attractive hedges against possible deterioration in the Turkish political environment. To clarify, we do not argue for a sure and sharp deterioration in the political scenario in Turkey, but would like to highlight that we see some material tail risks. What's more, as the markets are not pricing in the uncertainty that we see around the consensus base case scenario at the upcoming referendum, attractive and cheap hedges are available across the asset classes, in our view.
• Buy 5y Turkey CDS vs. Russia at flat spread
• Pay 1s3s in CCS curve (steepener)
• Buy 3-week EUR/TRY calls with strikes around 1.99
Strategy 1: We recommend tactically buying 5y Turkey CDS versus Russia at flat spread. Instead of buying 5y Turkey CDS outright, we implement this as a relative value trade in order to improve the carry. We chose to sell Russia CDS for the following reasons:
Macro-fundamentals momentum in favour of Russia. Our MDTI has detected a sharp fundamental improvement for Russia versus Turkey in recent months.
We illustrate a strong correlation between the 12-month rolling average of the MDTI index and Turkey minus Russia 5y CDS levels (normalised by dividing the differential by the average of the two CDS levels). The correlation broke down when the global crisis hit, raising concerns over the ability of Russian corporates to roll over external debt, while at the same time also causing a favourable correction in Turkey's external imbalances. Given that global liquidity conditions have improved, we expect this correlation to reassert itself.
Resilient oil prices should consolidate the past fundamental improvements. Analysing the impact of oil price changes on external accounts, fiscal balances and inflation for both Russia and Turkey, we notice an unsurprising divergence of trends. For an increase of US$10/bbl in oil prices, we would expect the current account to improve in Russia by +2% of GDP (Turkey -0.6% of GDP) and fiscal revenues to increase by 1.7% of GDP (with a marginal impact in Turkey). Inflation would be negatively affected in both countries, but much more in Turkey. In fact, according to our estimations, Turkish CPI would deteriorate by 1% over 12 months for each 10% of increase in the oil price, compared to a marginal 0.3% in Russia (see CEEMEA Macro Monitor, February 1, 2010). The Turkey to Russia CDS spread is trading in line with the current oil price and a sharp correction of commodities (not our base line scenario) would represent a medium-term risk to this trade. However, we present this strategy as a tactical way to hedge against a possible imminent increase of political noise in Turkey. As such, we do not see oil as a reason to refrain from opening this strategy.
Spread with bullish bias, but attractive valuations offer some cushion. Turkey typically has traded wide to Russia in an environment of low volatility and tight in periods of sharply rising volatility. Improved Turkish fundamentals and exposure of the Russian corporate sector to external debt make it unlikely that Turkey will trade at historically wide spread levels to Russia. However, we would expect the normalised spread to move 25-50% in favour of Russia - 40-80bp - over the coming 1-2 months. Risks: i) Russia's higher sensitivity to global credit conditions suggests that a sharp and sustained increase in global risk-aversion would adversely impact Russia versus Turkey; and ii) a sharp correction in oil price.
Strategy 2: In local rates, we recommended 1s3s CCS steepeners in last week's CEEMEA Investor. The curve has bull-flattened significantly over the past few weeks, on the back of persistent strong inflows. The market has priced out most of the rate hikes (as inflation looks set to fall towards end-2010) and significant amounts of risk premium was squeezed out of the curve (due to the government's fiscal prudence so far).
With 1s3s at around 70bp, the spread is close to the lows since March 2009. This implies that the curve is pricing in less than 75bp of hikes between 2012 and 2014, which compensates very little in terms of risk premium, in our view. Given that the CBT is still a long way away from hiking rates, we do not believe that the curve could flatten/invert further from here. We believe that the spread could widen by at least 30bp over the next few weeks, and this trade has a slightly positive carry/rolldown, at around 1-2bp per month.
Read Full Article »