On the week, benchmark Treasury yields fell 6-15bp and the curve steepened, which was a pain trade, as it seems that many investors had been moving into flatteners recently, a move that probably makes sense going into the FOMC meeting but didn't work in the most recent week. The old 2-year yield fell 15bp to 0.85%, 3-year 15bp to 1.41%, old 5-year 14bp to 2.29%, old 7-year 12bp to 2.96%, 10-year 9bp to 3.39%, and 30-year 6bp to 4.23%. Bill supply shrank much in October, which intensified the squeeze at the front end, with the 4-week yield down 2bp to 0.01% and 3-month 1bp to 0.05%. As a result of the paydown of almost all the SFP cash management bills, total bills outstanding fell by US$134 billion in October. Even with this big decline, however, it's hard to understand how bills can be so scarce when the increase over the past couple of years still amounts to about US$1 trillion. TIPS performed very well, mostly outperforming nominals even as a bit of a rebound in the dollar pressured commodity prices. As at the 10-year auction earlier in the month, the very strong demand at the 5-year TIPS auction apparently signaled a broader shift into the sector regardless of the short-term oil price moves that often drive relative performance on a day-to-day basis. The 5-year TIPS yield fell 21bp to 0.61%, 10-year 10bp to 1.38%, and 20-year 7bp to 2.02%. The 5-year TIPS yield has now been more than cut in half since mid-August, ending the week at its lowest level since July 2008, while the 10-year yield reached its lowest level since March. Thanks to a big day of outperformance Wednesday, when the prior backup in rates contributed to unusually low origination activity that was also seen in a big drop in the weekly mortgage applications survey, the MBS market about kept pace with the Treasury rally. This left MBS yields back down near 4.25% after they had risen to two-month highs approaching 4.5% Monday. The improved tone later in the week should keep 30-year conventional mortgage rates near the 5% level they've been averaging recently after a bit of a back-up from recent lows near 4.875% reached a few weeks back. Home sales will likely see some near-term downside if the first-time homebuyers' tax credit expires, but the current combination of mortgage rates and home prices still makes for unusually high levels of housing affordability that should be supportive going forward.
Risk markets took a big hit the past week to cumulate to a moderate correction from the recent highs. The S&P 500 lost 4% on the week for a 6% drop since the October 19 high. For all of October, the index was down 2%, obviously nothing too significant in itself, but it was the first down month since all the way back in February. In the latest week, high-beta financials plunged 7%, and the bounce in the dollar and corresponding pullbacks in commodity prices also contributed to poor performance by materials (-7%) and energy (-5%) stocks. But the weakness was very broadly based, and all major sectors saw significant losses. Credit markets also took a significant hit. In late trading Friday, the investment grade CDX index was 9bp wider at 109bp, the one mild positive in that being that it would only match the Wednesday close instead of going out at a new low in a month, like stocks. High yield traded more closely in line with equities. Through Thursday, the HY index was 28bp wider at 666bp, but a point-and-a-half drop Friday had it on pace of a more than 60bp widening on the week to the worst level since the beginning of October. Other markets also took major hits. The commercial mortgage CMBX market was down big, with the AAA off 5%, junior AAA 5%, AA 8% and A 5%. This left the AAA down about 2% on the month, but the reversal in the lower-rated indices has been much more pronounced after enormous rallies in September. The junior AAA surged from 47.15 on September 14 to a high of 60.60 on September 29, a 29% rally in a couple weeks, but at Friday's close it was down to 50.11, for a 17% pullback from the high. The subprime ABX market also was hit the past week, with the AAA index off 2.13 points (-6%) to 31.00. Recent prior performance had been strong, however, so this still represented a modest gain for the month.
Real GDP rose at a 3.5% annual rate in 3Q, ending a run of four-straight declines (and five of the prior six quarters). A slower, but in absolute terms still very intense, pace of inventory liquidation added 0.9pp to growth, while net exports subtracted 0.5pp as imports (+16.4%) rebounded a bit more sharply than exports (+14.7%). Final domestic demand gained 3.0%, paced by a 3.4% gain in consumption after a near-record drop in the year through 2Q and a 23.4% rise in residential investment after 14-straight declines. The cash-for-clunkers-driven surge in auto sales accounted for much of the consumption upside, but ex autos spending gained a solid 1.8%, highest in two years. Business investment (-2.5%) on the other hand remained soft on weakness in structures. Cash-for-clunkers payback in consumption and not as big a boost from a further slowing in the rate of inventory destocking should lead to a more moderate pace of growth in 4Q, but a better trajectory to ex auto retail sales - which posted a real gain of 0.3% in September on top of a 0.55 rise in August even as a correction in auto sales caused overall real spending to fall 0.6% - points to a somewhat more muted slowdown than we thought a month ago. 4Q growth at this early point appears to be on track for a gain near +2.5%, which would leave annualized 2H growth at +3%. We continue to look for this muted rebound to extend into 2010, when we forecast 3.2% GDP growth on a 4Q/4Q basis.
Against the upside in GDP, the week's other two key releases - consumer confidence and new home sales - were weaker. The Conference Board measure of consumer confidence fell 6 points in October to 47.7 as pessimism about the labor market reached extreme levels. The University of Michigan index was also weaker, falling to 70.6 from 73.5 (though this was revised up from the early October reading of 69.4). In the Conference Board survey, the current conditions index fell 1.3 points to 20.7, a low since early 1983 and not far from an all-time low. The percentage of respondents describing jobs as plentiful fell a couple of tenths to just 3.4%, while the percentage describing jobs as hard to get rose nearly three points to 49.6%, the worst net view since May 1983 when the unemployment rate was 10.1%, a level that was likely nearly reached this month. New home sales fell 3.6% in September to a 402,000 unit annual rate after a huge prior rebound off the record low hit in January. Even after this pullback, however, sales are up 22% from that early year trough. With single-family home completions hitting a record low in September as they continue to catch up with the prior collapse in starts, inventories of unsold homes continued to fall even with the drop in sales, declining 4% to 251,000 units, a low since 1982.
The upcoming week has a very busy calendar, with the FOMC meeting, the key early round of data releases for October highlighted by the employment report Friday, and the Treasury's quarterly refunding announcement on Wednesday. The Fed is in an interesting position at the moment. While the end of the recession means that it is clearly time to start thinking about exit strategies, the reality on the ground is that quantitative easing is in the process of a big further ramp-up. Because most of the emergency liquidity injections (including TAF and related lending through foreign central banks, CPFF, PDCF and discount window borrowing) have largely unwound on their own as market conditions have settled down, there has recently been only a limited offset to the Fed's ongoing mortgage and agency purchase programs. So excess reserves have recently been surging to new highs after having been roughly stable for most of the year, and this will likely extend into next year. Excess reserves were a bit below US$800 billion at the end of August before rising to US$987 billion in mid-October. This Thursday's report should show a move above US$1 trillion, and we expect a peak near US$1.2 trillion early next year, for about a 50% further increase in a six-month period. As this is happening, however, certainly the Fed is already thinking carefully about when and how to bring this back down, and so heading into the FOMC meeting, it's clear that a change in the wording of the official statement is very much in play. Our base case at this point is that the Fed begins reverse repos to drain excess reserves in March, as the MBS and agency purchase programs are wrapping up. In our view, it is a close call on whether the FOMC decides to adjust the "extended period" wording at this meeting to start to move a bit rhetorically towards this shift or waits until the December session to do so. The recent data flow has been somewhat mixed, but we suspect that the solid GDP report together with broader indications that a subpar but sustainable economic recovery is starting to unfold tips the scale toward the likelihood that a change will occur this time round, though if markets remain under pressure in coming days, that could certainly keep policymakers from doing anything just yet. We expect any tweaks they do make to the key policy language to be quite minor at this point. We're thinking that the change might involve something along the lines of substituting "a highly accommodative monetary policy" for "exceptionally low levels of the federal funds rate".
Ahead of Wednesday's FOMC announcement, the Treasury will give its quarterly refunding announcement. We look for a US$82 billion refunding package consisting of US$40 billion 3s, US$25 billion 10s and US$17 billion 30s, which would be an increase of another US$1 billion for the 3-year and US$2 billion for the 10-year and 30-year. We also expect the Treasury to confirm a shift from 20-year to 30-year TIPS starting January. The debt managers appear determined to reverse the shortening in the average maturity of the outstanding Treasury debt that has occurred over the past couple of years in an expeditious manner. We believe that they would probably like to boost the average maturity of the Treasury debt outstanding from its current unusually low level of about 4 years up to an elevated 6-7 years. To accomplish such a substantial shift in the average maturity of the stock of outstanding debt over a reasonable timeframe will require a much bigger extension in the average maturity of the flow of new debt being sold. So, we expect the Treasury to continue paying down bills over the next year, and we expect the upcoming auction announcement to see a continued move higher in longer-maturity issue sizes. So, even though overall Treasury issuance will likely be a lot lower in fiscal 2010 than fiscal 2009, the duration of the issuance and overall gross coupon sales will likely be much higher.
In addition to the employment report on Friday, key data releases due out include ISM and construction spending Monday, factory orders and motor vehicle sales Tuesday, non-manufacturing ISM Wednesday, and productivity and chain store sales Thursday:
* The regional reports for October were mixed (3 up and 3 down). However, a couple of those posting increases - Empire and Chicago - were up a lot. So, we look for a small rise in the national ISM index relative to 53.0 from 52.6 in September. In particular, we look for a rebound in both orders and production along with a continued climb in the inventory index (which is actually reflective of a slower pace of decline). Finally, the price gauge is expected to tick up to 65 (versus 63 in September) as a result of higher quotes for energy-related items.
* We look for a 0.2% dip in September construction spending following an uptick in August. In particular, residential activity is expected to flatten out, reflecting the fact that the volume of houses under construction is still declining even though new starts appear to have bottomed. Meanwhile, the non-residential category appears poised for some further significant declines, given widespread stress in the commercial real estate sector. Finally, we look for a modest uptick in the public category, although there is scant evidence that infrastructure projects funded by the fiscal stimulus legislation enacted back in February are having any meaningful impact on the data.
* We expect factory orders to rise 0.6% in September. The durable goods component of orders showed a 1% rise, but some price-related softness in the non-durables category should help to shave the growth in overall factory bookings. Meanwhile, shipments should also show a modest gain (+0.5%), while inventories continue to slide (-0.8%). This implies another slight dip in the I/S ratio (to 1.37 from 1.38).
* Industry surveys point to a solid rebound in motor vehicle sales in October, so we look for an improvement to 10.4 million units annualized from the subdued 9.2 million unit pace seen in September, which obviously reflected a cash-for-clunkers payback. The recent sales rebound largely reflects greater availability of some popular models that were in very short supply in the immediate aftermath of C4C. Note that even with the recent sales boost tied to the success of C4C, sales on a year-to-date basis are down 25% from 2008.
* We forecast a 6.6% gain in 3Q productivity and 4.5% drop in unit labor costs. The combination of a solid gain in output (+4.0%) and a further decline in labor input (-2.6%) implies another very sharp jump in productivity. Indeed, if our estimate for 3Q is close to the mark, the productivity growth seen in the past two quarters would be the best since 2003 and the third best in the last 25 years. Moreover, the year-on-year growth rate should jump all the way from +1.9% to +3.6%. Meanwhile, unit labor costs are likely to post a sizeable decline for the third consecutive quarter, and the year-on-year rate should drop all the way from -1.2% to -3.4%. This reflects the unusually aggressive cost control that has been helping to sustain corporate profitability.
* We look for a 150,000 decline in October non-farm payrolls. The downtrend in jobless claims points to a likely moderation in the pace of payroll decline. Moreover, factoring in BLS's preliminary estimate of the 2009 benchmark revision, it appears that the level of private employment is below the July 2003 trough - even though the economy's output is now 10% higher in real terms. This performance reflects strong gains in productivity over the past several years, but such gains can only be stretched so far. As demand begins to firm, companies will need to add some labor in order to grow top-line revenues, and we believe that the labor market is in the process of transitioning to just such an environment. In October, we look for a slower pace of job loss in sectors such as construction, retail trade and government. Meanwhile, the steady climb towards a 10% unemployment rate is expected to continue. However, labor force participation appears to be drifting lower (after having risen during the first half of the year), and this is helping to temper the pace of increase in the unemployment rate, so we look for only a further 0.1pp uptick in October to 9.9%. Finally, we look for a flat workweek and only a very slight uptick in the average wage rate.
Executive Summary
We conduct scenario analysis in order to investigate fiscal sustainability for Greece and Ireland. Although both will likely have similar gross debt/GDP ratios for 2009 (110%) and 2010 (118%), on our estimates, there are meaningful differences in the underlying fiscal picture.
The most relevant differences are along two dimensions: (1) The measurement of the debt; and (2) The skew of the risk profile for the fiscal outlook.
(1) Which debt? Comparing gross debt can be misleading:
• For Ireland, there is a meaningful difference between gross and net debt. Ireland has a significant amount of assets in the National Pensions Reserve Fund (12% of GDP as of September 2009); in addition, government cash balances are substantial (another 15% of GDP). Net debt/GDP for Ireland - the more appropriate measure of a government's fiscal position - is therefore about 27% lower than gross debt/GDP: 70% for 2009 and 90% for 2010, on our estimates.
• Against a substantial part of the debt (gross and net), the Irish government has acquired assets - albeit of uncertain value. Unless the economic outturn is particularly bad (as is handicapped in our scenarios below), these assets will offer at least some payoff in the future.
(2) Skewness of the risk profile - Ireland more risky in short term, Greece more risky in long term: The risks surrounding the Irish fiscal position are mostly short-to-medium term; they emanate mainly from the possibility that the deep recession-cum-banking crisis would result in chronic weakness of the economy or even morph into a multi-year deflationary spiral. If the economy navigates this threat successfully, we reckon the fiscal outlook would improve substantially, albeit possibly still requiring additional fiscal efforts in the short term.
The risks for Greece are more skewed towards the medium-to-long term. A decidedly unfavourable long-term demographic outlook implies that Greece only has a narrow window of opportunity to reduce government indebtedness and implement the reforms that would allow faster growth during the demographic transition.
One simplified way to show the long-term risks is by comparing the primary balance (the budget deficit excluding interest payments) required to keep the debt from rising in our scenarios. The scenarios are not the same for the two countries and so their comparability is limited; however, a consistent result emerges: Greece requires higher primary surpluses (a negative deficit is a surplus) for debt stabilisation in all scenarios. This is in part due to the adverse long-term demographic projections.
A different way of looking at the same issue would be to ask which economy is, in the medium term, likely to generate the economic growth that would erode the debt. In other words, which economy is the more dynamic one? On the basis of past performance as well as current institutional reality, we believe that the answer is Ireland - provided the economy recovers reasonably quickly from the deflationary slump.
1. Fiscal Sustainability Scenarios
Given that interest payments on the outstanding debt are fixed, deficit and debt reduction require that the non-interest part of the government budget - the primary balance - be in surplus (see Budget Terms Glossary below). In what follows, we will therefore focus on the primary surplus as the fiscal authority's sole decision variable (implicitly assuming that defaulting on interest payments is not an option either government would consider).
In our benchmark analysis for both countries, we arbitrarily assume a primary surplus of 0.7% of GDP, maintained from 2013 until 2050, the end of the scenario horizon. If this primary surplus achieves a stable (i.e., non-increasing) debt/GDP ratio towards the end of the scenario horizon, we will conclude that fiscal policy is sustainable. If the debt path is increasing, we will conclude that fiscal policy is unsustainable. Our numbers include ageing costs as projected by the European Commission's working group on ageing populations (AWG). We emphasise that our scenarios do not have a bull-base-bear character. Rather, the attempt is to handicap the relevant risks for each country's fiscal outlook - both on the upside and the downside. Finally, because of the discrepancy between net and gross debt for Ireland discussed above, we will contrast net debt for Ireland with gross debt for Greece (as a share of the respective GDP).
Budget Terms Glossary
Primary expenditure: All budget expenditure items except interest payments on outstanding debt.
Primary balance: Primary expenditure less total revenue. If primary expenditure exceeds total revenue, the primary balance is in deficit (primary deficit). If primary expenditure is less than total revenue, the primary balance is in surplus (primary surplus).
Net borrowing (general government deficit): Excess of total expenditure over total revenue. If total expenditure exceeds total revenue, net borrowing is positive and the budget is in deficit.
Net borrowing versus primary deficit: By definition, the primary deficit is the general government deficit (net borrowing) excluding interest payments on the outstanding debt. Put differently, net borrowing is the sum of interest payments and primary deficit:
1.1 Greece
The focus of our scenarios for Greece is the supply side of the economy. In the long run, GDP growth will likely be constrained by labour force growth - which in turn depends on population growth. Greece's population and hence the labour force is projected to decline over the next four to five decades. In the medium term, structural reforms in labour and product markets could boost employment and growth (as in our ‘Continued Convergence' scenario below). Conversely, factors that make labour and product markets more rigid will restrain employment and overall economic growth (as in our ‘Lost Decade' scenario).
1. ‘Continued Convergence' scenario: Here, it is assumed that broad structural reforms increase GDP growth and lower unemployment in the transition period as the economy converges to a new equilibrium with a lower labour force (see Appendix A in the full report for numerical assumptions of all scenarios). The results, for a variety of interest rate levels, are illustrated. Appendix C in the full report shows the primary surplus levels for which the debt ratio is stabilised, depending on the interest rate.
• In the baseline case of a 5% implicit interest rate on the debt, the debt ratio at the end of the forecast horizon is 158% and rising. Higher interest rate levels lead to even higher debt ratios: fiscal policy is not sustainable.
• Until roughly the middle of the scenario horizon, the debt ratio is actually decreasing or stable (except for the case of a 7% interest rate). The fact that it increases thereafter testifies to the strength of the demographic headwinds Greece faces.
• Depending on the (implicit) interest rate on the debt, the primary surplus that stabilises the debt/GDP ratio is 2.1-4.0% of GDP (see Appendix C in the full report).
• The level at which the debt ratio is eventually stabilised also increases with the interest rate on the debt. In this scenario, the debt/GDP ratio is stabilised at 88-96% of GDP, depending on the interest rate.
• Debt stabilisation is achievable: primary surpluses of the magnitude required have been attained in the recent past on a sustained basis.
• However, stabilising the debt would require substantial additional fiscal measures: according to our back-of-the-envelope calculations, for the debt-stabilising 2.1% of GDP primary surplus to be achieved by 2013 necessitates a permanent fiscal consolidation of at least 10% of GDP over the next four years.
2. ‘Baseline IMF' scenario: We assume that the economy evolves according to the baseline scenario in IMF (2009), with no deviations.
• Given the benchmark assumption of a 0.7% of GDP primary surplus, fiscal policy again turns out to be unsustainable in this scenario. Even for a 5% implicit interest rate, the terminal debt ratio is very high (at 224%) and rising.
• The debt-stabilising primary surplus (as a share of GDP) is 2.6-4.7% for the range of interest rates on the debt we consider here. Such primary surpluses would stabilise the debt ratio at 107-113% of GDP.
• The highest primary surplus/GDP ratio on Eurostat records for Greece was 4.6% in 1998: high interest rates may well require unprecedented primary surplus levels.
3. ‘Lost Decade' scenario: We assume that the economy ‘loses' the coming decade to social unrest and poor labour relations; this acts as a negative supply shock which lowers growth but increases inflation. Structural reforms are enacted after 2020 and bear fruits only later.
• Once again, fiscal policy is not sustainable: debt/GDP is high and increasing at the end of the scenario horizon (218% in 2050 for a 5% interest rate).
• Stabilising the debt ratio in this scenario requires, depending on the interest rate on the debt, a primary surplus/GDP ratio of at least 2.4% (4.7% for a 7% implicit interest rate on debt). Debt would be stabilised at levels exceeding 118% (see Appendix C in the full report).
Summary
Fiscal sustainability in Greece requires substantial fiscal consolidation over the coming years. As such, the primary surplus levels (as a percentage of GDP) required to keep the debt ratio from rising have been achieved in the past. However, Greece will need to show a consistency in its budgetary policies it has not shown before (recall that the debt-stabilising primary surplus will have to be maintained indefinitely for debt stabilisation). The upcoming demographic transition and its economic impact through slower growth and higher health and pensions expenditure mean that Greece has very little margin for error if it wants to ensure fiscal sustainability.
1.2 Ireland
An assessment of Ireland's fiscal outlook is complicated by the fact that much of the debt is a result of the government acquiring impaired assets from the banking sector (see Scenarios on Irish Fiscal Sustainability, July 21, 2009, for more details). The value of these assets is difficult to establish at this point, and will ultimately depend on the performance of the Irish housing and commercial real estate sectors over the medium term. This, in turn, will rest crucially on the prevailing macroeconomic conditions, particularly on whether and how soon the economy will be able to escape the current deflationary slump.
Read Full Article »