On the week, benchmark Treasury yields fell 6-12bp after rallying 10-19bp Friday to more than reverse modest losses posted through Thursday. The 2-year yield declined 6bp to 0.72%, 3-year 9bp to 1.15%, 5-year 12bp to 1.98%, 7-year 11bp to 2.64%, 10-year 10bp to 3.19% and 30-year 10bp to 4.12%. TIPS breakevens were similarly seeing some upside over the course of the week through Thursday but reversed back down hard in Friday's renewed fear trade that included big losses in commodity prices and a surge in the dollar, with July oil losing 4% on the week and the LME's base metals composite - which was soft all week on China slowdown fears even before Friday's global rout - 11%, while the dollar index surged 2%. The dollar's biggest gain on the week was actually against the Australian dollar, which sank to $0.824 from $0.847, but the more headline-grabbing move clearly was the drop in the euro to another four-year low of $1.198 from $1.227, with two-thirds of that drop on Friday. On the week, the 5-year TIPS yield fell 6bp to 0.29%, 10-year 4bp to 1.22% and 30-year 6bp to 1.75%. This lowered the 10-year inflation breakeven 4bp for the week to 1.98% after an 11bp decline Friday. Interest rate volatility actually didn't move much during Friday's big move in yields and ended the week little changed, which was supportive of another robust week for the MBS market. Fannie 4% MBS was trading just below par at Friday's close, sending current coupon mortgage yields down to just above 4% after a small outperformance versus the week's Treasury market rally. Average 30-year mortgage rates have been close to 4.8% the past three weeks, just above the record low of 4.71% reached in December, and if continued European fears in coming days allow Friday's rally to be sustained into the coming week, we could see a new low shortly. Mortgage refinancings have been rising quite sharply recently, providing some rate stimulus to consumers from this financial crisis on top of some real income stimulus that has been provided by the big pullback in gasoline prices recently.
After rising steadily from mid-March until late May, with the rate of increase accelerating sharply for a few weeks in May, 3-month Libor has been just about unchanged now for eight straight days, fixing Friday at 0.537%, effectively unchanged on the week. This has left the spot 3-month Libor/OIS spread at 30 or 31bp consistently over this period. While this is up 20bp from mid-April, we are not overly concerned considering that when the 2007-09 financial crisis first broke out in August 2007, the spot Libor/OIS spread was through 30bp immediately, was through 60bp within a week, was near 75bp at the end of August, and was moving towards 100bp by mid-September. After that initial August 2007 blowup, the spread didn't even get back through 30bp again until July 2009, so if that's as bad as things get in dollar funding markets in this crisis, it's hard to see it as much of an issue. What had been becoming increasingly more concerning than this spot spread a couple weeks ago was how much the market was pricing it would escalate in a fairly short period of time, apparently seeing some risk of an August/September 2007 situation developing over the summer even if the stresses are nowhere near that high now. But while there was a bit of backtracking in the general renewed fear trade Friday, on the week there was still a modest further scaling back of these forward-funding worries. As long as this key potential liquidity transmission mechanism remains contained, systemic spillover worries from the European situation eventually should calm again down to some extent, we would expect. On the week, the forward Libor/OIS spread to September fell to 52bp from 55bp and the May 24 high of 71bp as the Sep 10 eurodollar contract rallied 5bp to 0.795%.
This was the second week in a row where risk markets tried to set aside European worries to post gains into Thursday only to be knocked back down by a renewal of fears Friday. But there was a much worse downturn this Friday from the combined impact of the sudden situation in Hungary, steady widening in Spain/Germany spreads, and heavy pressure on European financials than there was the prior Friday from the ‘tape bomb' of Fitch's downgrade of Spain, which in retrospect was really immaterial. On the week, the S&P 500 lost 2.2% after plunging 3.4% Friday. In contrast to the stability in rates, equity market volatility shot up, with the VIX moving up to 35.5 from 32.1. Credit was mixed, with investment grade performing quite poorly but high yield managing to outperform a bit on the week by not doing all that bad in the sell-off after having lagged while other risk markets had been doing better through Thursday. The IG CDX index widened 9bp to 126bp, while the HY CDX index was about 40bp wider near 665bp. Subprime and commercial real estate also held up better in Friday's sell-off after having done well through Thursday to end up with decent gains on the week - the AAA ABX index rose 2%, AAA CMBX 1%, junior AAA CMBX 2% and AA CMBX 4%. The gradual move wider seen all week in peripheral European government debt spreads over Germany - which by Friday had the Spain/Germany 2-year spread near 230bp, not far from the early May peak of 240bp as Spain starts to look ahead to a heavy schedule of July debt redemptions - applied spillover pressure onto the muni bond MCDX market all week, and this intensified in Friday's big fear trade. The 5-year MCDX index was up to 185bp Friday afternoon, 20bp wider on the week and through the prior wide for the year of 180bp reached in February.
The past week's economic data showed broad upside, including important underlying details of the employment report, notwithstanding the disappointing headline payroll number. The key run of initial data for May - both ISM surveys, early indications of consumer in the auto and chain store sales results, and the robust hours and earnings numbers in the employment report - plus the biggest gain in construction spending in April in a decade indicated that the acceleration in the economy that became increasingly apparent in the underlying resilience in the February data despite the severe weather has extended well into 2Q. Growth in 2Q looks set for a decent acceleration from the +3% recorded in 1Q, likely running near +4% overall and with much more robust underlying details than in 1Q, with the inventory contribution in 2Q likely to flatten out at least temporarily after accounting for half of 1Q's 3% growth.
Non-farm payrolls surged 431,000 in May, but almost all of the gain reflected 411,000 temporary census hires (which will start to reverse in June). Private sector job growth slowed to 41,000 after a 218,000 rise in April, with the volatility probably largely driven by the unusually good weather in April. Weather can be quite important for jobs in April as seasonal industries ramp up, and the impact seemed apparent in swings in outdoor oriented sectors like construction - down 35,000 in May after a 14,000 rise in April - and leisure - up only 2,000 in May after a 35,000 jump in April. Beneath this short-term weather-related volatility in payrolls, other key details of the report were strong. The unemployment rate fell to 9.7% from 9.9%, but probably mostly from census impacts. Hours and earnings numbers, which are private sector only, were also quite robust, however. The average workweek rose a tenth to 34.2 hours, high since November 2008, which combined with the gain in private sector payroll lifted total private sector hours worked by a solid 0.3%. With average hourly earnings accelerating to +0.3%, aggregate earnings - an important proxy for total wage and salary income - soared 0.7% on top of a 0.6% gain in April and 0.4% rise in March. Note that with the recent seasonally unusual softness in gasoline prices, inflation will be weak in May, so that +0.7% number will be near +1% in real terms, and the +7% annualized surge in nominal aggregate payrolls over the past three months will be about the same, and a lot more impressive, after adjusting for flat prices in the three months through May. Within the upside in overall hours worked, the manufacturing sector continued to lead. Factory sector jobs rose 29,000, the average workweek jumped to 40.5 hours from 40.2, and aggregate hours worked jumped 1.1% - pointing to another very strong industrial production report in May.
That factory sector strength in May had been previously seen in a surprisingly robust manufacturing ISM report after broad moderation in the regional surveys. The composite manufacturing ISM index fell only fractionally to 59.7 in May from the six-year high of 60.4 hit in April, and underlying details were stronger, with all the pullback accounted for a drop in the inventories index, while orders (65.7 versus 65.7) and production (66.6 versus 66.9) held steady at unusually high levels and employment (59.8 versus 58.5) posted a further gain. In addition to the disconnect from the main regional reports, however, some details suggested that this strength may have, to some extent, reflected a lag in fully reflecting recent developments. In particular, the prices paid index was little changed at a very high level despite the pullback in commodity prices over the past month, and the export index hit a 22-year high, which was surprising given the turmoil in Europe. Meanwhile, the composite non-manufacturing ISM index was steady at 55.4 in May, a high since 2006. The breadth of the expansion improved though, with a record 16 of 18 sectors reporting growth, up from 13 in April and March. The most notable shift among the components was a one-point gain in the employment index to 50.4, the first reading above the 50-breakeven level since December 2007. The business activity gauge (61.1 versus 60.3) moved to a five-year high, but the orders index (57.1 versus 58.2) moderated a bit.
In the upcoming week, investors seem likely to continue Friday's renewed focus on the European debt distress after the brief period of relief we had this past week. The US calendar is fairly light, with Treasury market focus likely to be largely on supply until Friday's retail sales report. The Treasury will auction $36 billion 3s, $20 billion 10s and $13 billion 30s on Tuesday, Wednesday and Thursday. The 3-year size was cut $2 billion for a second-straight month, but the 10-year and 30-year reopening sizes were held steady, as the debt managers are extending the average maturity of coupon issuance even as they trim back overall coupon issuance modestly to allow a slowing in the rate of T-bill sector paydowns as the deficit starts to narrow. Other than the retail sales report the most notable data releases are the trade balance and Treasury budget Thursday:
* We look for a slight narrowing in the trade deficit in May to $40.0 billion, with both exports (+1.6%) and imports (+1.0%) expected to post decent gains. Export upside should be led by capital goods, with industry figures pointing to a sharp rebound in the volatile aircraft sector while factory shipment figures show strength in high-tech industries. Higher prices should also support upside in industrial materials. On the import side, higher prices and volumes point to some upside in petroleum products, while port data suggest that imports of other goods should also rise.
* We expect the federal government's budget deficit to narrow $57 billion from a year ago in May to $133 billion. The swing reflects a number of factors, including: the special one-time $250 checks that were distributed to social security beneficiaries in May 2009 ($13 billion), lower TARP outlays ($12 billion), a calendar effect that accelerated some regular monthly payments into April (about $10 billion), and a quicker fall-off in tax refunds during the current filing season ($9 billion). We continue to see the deficit tracking at $1.25 trillion (or -$8.5% of GDP) for the fiscal year as a whole.
* We look for a 0.4% rise in overall May retail sales and a 0.3% decline ex autos. We see a number of cross-currents in the May sales data. First, due to differing seasonal factors, the modest rise in unit sales of motor vehicles appears to translate into a sizeable advance in the auto dealer component of retail sales (+3.5%). Second, declining prices should lead to a significant drop in gas station receipts (-2.2%). Third, sales at building materials outlets soared in recent months and much of this upside appeared to be attributable to the impact of the ‘cash for appliances' programs that were quite popular in many states. Programs in most states wound down in April. So, we look for a significant pullback in this sector (-4.0%). The key item to watch in this report is retail control. And, based on the mildly positive chain store results, we look for control to rise 0.4%.
Summary and Conclusions
Like the other major economies, the euro area is now expanding again. However, the looming fiscal tightening is a potential threat to the fragile recovery. In this report, we take a close look at the current and prospective fiscal austerity programmes of the euro area member countries. We conclude that the overall fiscal tightening of 0.6% of GDP this year and up to 1.5% next year would reduce euro area GDP by 0.4% this year and up to 1% the next. However, the euro depreciation will likely offset in full the negative impact of fiscal austerity.
The effects will differ in each euro area country. We recently revised up our GDP growth forecasts for Germany, Italy and Portugal (2010), down for Spain, Greece, France and Portugal (2011). For the euro area as a whole, we maintain our below-consensus forecast of 0.9% this year and 1.1% the next. But despite no change in the overall GDP numbers, domestic demand might be even more sluggish, and exports stronger - courtesy of the combined effect of a weaker currency and further belt-tightening.
What Has Already Changed in the Fiscal Landscape?
There is virtually no correlation between the budget balance and GDP growth. So changes in the fiscal policy stance don't matter in the euro area? We think that they do matter a great deal, but not necessarily in a mechanistic sense. Indeed, measures of public sector activity - ranging from government consumption to the broader aggregate of government spending (which includes transfers) and public investment - show a meaningful link with GDP growth.
The Public Sector - Is it a Key Economic Driver?
Government spending is a fairly small share of euro area GDP relative to, say, household spending or total investment. Indeed, in 2009 it accounted for approximately 21.5% of GDP in the euro area, up from 19.8% at the beginning of the decade. This compares with a weight of 58.2% for household spending last year, and 19.4% for total investment. (Exports and imports of goods and services account for 41% and 40.3% of euro area GDP, respectively, so the share of net exports - i.e., exports minus imports - is negligible.)
However, the relevance of government spending - and the role of the public sector more broadly - as an economic driver, either directly or indirectly, should not be underestimated. Indeed, government spending has always contributed positively to GDP growth over the past decade, unlike household consumption, investment and net trade. Since 2000, government spending has accounted for about one-third of euro area annual GDP growth of 1.4%, on average. Clearly, this looks set to change. With a more stringent public purse in many euro area countries, the chances are that government spending might even contribute negatively to GDP growth over the next couple of years. What's more, the public sector does affect the economy not only through changes in government spending. Other channels, from public investment to tax policy, are also important.
Direct and Indirect Effects - Are They Important?
Empirical evidence confirms that government involvement in economic activity does play a key role and can affect GDP growth - though in a counterintuitive way, in some instances. We explore various public sector-related indicators and look at their correlation with GDP growth for the euro area as a whole and its constituencies:
• Government consumption, i.e., government spending which buys goods and services produced in the economy and which is not a transfer payment of money collected in taxation from one group in society to another. Government consumption counts towards GDP, while transfer payments take money from some people and give it to others. Most day-to-day health and education expenditure count as government consumption. Building new hospitals is government investment; old age pensions are a transfer payment.
• Public investment, i.e., gross fixed capital formation of the general government - including, among other things, structures and equipment used by the military, which are similar to those utilised by civilian producers, such as docks, airfields, roads and hospitals; light weapons and armoured vehicles used by non-military units. The purchase of military weapons and their supporting systems is still part of intermediate consumption and not included in gross fixed capital formation.
• Government spending, i.e., the sum of intermediate consumption, public investments and payments for compensation of employees, subsidies, social benefits and transfers, and tax-related disbursements, among other things.
Looking at the past 20 years or so, the evidence suggests that government consumption, and, to a lesser degree, public investment are positively correlated with economic growth.
Conversely, overall government spending is negatively correlated with economic growth. Although this may seem counterintuitive at first sight, one explanation could be that, in periods of weak economic activity or recessions, government spending tends to increase - courtesy of the functioning of the so-called automatic stabilisers (such as unemployment benefits), which are quite substantial in several euro area countries.
Similarly, economic growth is adversely affected by an increase in government revenue (i.e., receivables on taxes, transfers, subsidies and contributions, as well as other sources - such as privatisation receipts - to help finance spending) as a share of GDP. For example, a tax hike will exert a negative impact on growth - all else being equal. In this case, too, the inverse correlation seems fairly high - at least over the 1992-2009 period under examination - just like with government spending.
Austerity Underway - Not Just in the Iberian Economies
So, changes in the fiscal policy levers do have a non-negligible impact on economic activity. With all countries in the euro area set to cut their budget deficits over the next couple of years, this will undoubtedly exert a negative effect on GDP growth. However, not all will implement their belt-tightening programmes simultaneously. Apart from Ireland (the first mover on the fiscal tightening scene back in 2008), some have already started. After Greece - see Our First Assessment of Greece's Loan and Austerity Package, May 3, 2010 - it is now the turn of Spain, Portugal and Italy. Others are likely to follow shortly, we think (France, and - perhaps further down the line - the Netherlands and Belgium). Some (notably Germany) will use their extra fiscal leeway to implement a more gradual adjustment programme. In particular, only the so-called EMU periphery has tightened (or is about to tighten) the belt this year - at least so far:
• In Spain, the new extra savings amount to €15 billion in 2010-11. This is equivalent to about 0.5% of GDP this year and 1% next year. Measures include: abolition of ‘birth payment' of €2,500 per household, 5% cut in civil servants' wages this year and a wage freeze next year, elimination of 13,000 public sector jobs, suspension of index-linked pensions in 2011, and spending cuts at the regional level. Factoring in the announced savings into our forecasts, we now expect the budget deficit at 9.4% of GDP this year and 6.7% the next (see Spain Economics: Extra Fiscal Tightening - In Small Steps, May 12, 2010). What's more, further fiscal austerity measures look likely, in our view, on three fronts: 1) a new wealth tax will likely apply to individuals with net worth greater than €1 million - subject to parliamentary approval; 2) further income tax rate hikes seem on the cards; and 3) a ban on local councils from issuing long-term debt - previously planned for this year - will take effect in 2011.
• In Portugal, the new extra savings amount to approximately 1.2% of GDP in 2010 and 2.2% in 2011. Apart from bringing forward the phasing-out of all the anti-crisis stimulus measures and some tightening planned for next year, measures include: cuts in central government spending, lower transfers to local governments and state-owned enterprises, 5% wage reduction to holders of political and public management offices, introduction of motorway tolls, increase of 1pp up to the 3rd income bracket and an increase of 1.5pp for the 4th income bracket onwards of personal income tax, and a 2.5pp increase in the corporate income tax for taxable profits exceeding €2 million. Factoring in the announced savings into our forecasts, we now expect the budget deficit at 7.5% of GDP this year and 4.7% the next.
• In Italy, the Combined Report on the Economy and Public Finance (known as RUEF in Italian), published in May, had already mentioned that the government would have needed to tighten its primary budget by about 1.6% of GDP over 2011-12, in order to meet the fiscal targets mentioned in Italy's latest stability programme submitted to the European Commission - to which it reiterated full commitment. In its present form, the resulting €24 billion package will include: 1) a three-year freeze on state wages; 2) pay cut for top public sector workers; 3) extra measures to curb tax evasion; and 4) delayed retirement for those due to retire next year. Factoring in an extra belt-tightening of the above-mentioned magnitude into our forecasts, we now expect the budget deficit at 5.1% of GDP this year and 3.7% the next.
How Much More Tightening Is Needed?
It all depends on what ‘how much' really means. Given the very different starting positions, the amount of consolidation required will vary significantly by country. The IMF has estimated the adjustment needed between 2010 and 2020 to bring the debt below 60% of GDP by 2030. According to these estimates, the cumulative required adjustment varies from around 4% of GDP in Germany and Italy to about 8% in France and Portugal, and 9% in Greece and Spain (see Fiscal Monitor - World Economic and Financial Surveys, May 14, 2010). And the variation in the adjustment needed to meet the Stability and Growth Pact's 3% ceiling for government borrowing might be even larger, as suggested by the annual reduction in the structural primary deficit that the European Commission believes is necessary to bring the unadjusted deficit below 3% of GDP before its deadlines.
Where Might the Extra Squeeze Come From?
The various EMU members will have different preferences in terms of the spending categories that could eventually be trimmed. Ireland spends more than average on healthcare, education (together with Portugal) and environment, housing and recreation, Italy on general public services, Greece on defence and economic affairs, Germany on social protection, and Spain on public order and safety.
What's more, further job reductions in the public sector might be considered in some countries, as was the case in Spain and Greece recently. Alternatively, there might be further hiring freezes in the public sector, or the introduction of, say, 2:1 rules of retirements/recruitments. On this front, France is the country with the highest share of workers in the public sector (36%); at the other end of the range there's Portugal (25%).
Post-Election Tightening in the Benelux Countries...
Fiscal consolidation has so far taken a backseat in the Netherlands and Belgium. This postponement reflects, at least in part, the fall of both governments in February and April, respectively - along with the difficulty, in both countries, to form a new ruling coalition. In particular:
• In the Netherlands, the election campaign is still in full swing, with the vote scheduled for June 9. Traditionally, fiscal policy has ranked quite high among the various governments' priorities, and the country has sound - and well deserved - credentials in terms of fiscal consolidation. Although the situation remains fluid on the political front and the magnitude of an eventual tightening may depend on the outcome of the election, a report by the Netherlands Bureau for Economic Policy Analysis (known as CPB in Dutch) shows that the political parties' programmes encompass cutbacks in government spending and, in most cases, increases in the tax burden.
• In Belgium, the vote is scheduled for June 13. Although the budget deficit is not particularly wide (around 6% of GDP in 2009), a triple-digit debt/GDP ratio of approximately 100% might well translate into market worries from a long-term fiscal sustainability standpoint. The election outcome will likely affect the fiscal policy outlook, at least to some degree, in this country too. But this will have to do mainly with the intensity of the belt-tightening, which looks very much on the cards regardless of the outcome of the election.
...While Germany Can Afford to Wait a Little Longer
Being in a stronger fiscal position relative to most European countries - and courtesy of its solid reputation as a fiscally responsible country - Germany has more time to cut its relatively contained budget deficit. And its benchmark status in euro area bond markets does imply more leeway on the fiscal front. But the fiscal policy stance is shifting in Germany too, we think. Not only has the government already postponed its previously announced tax cuts, it is about to hold an offsite (ending on June 7) to discuss budget savings for next year (probably around €10 billion).
France?
Harsh fiscal austerity doesn't seem to be on the cards - at least not at this stage. This sets France apart from virtually all the other euro area members. But its fiscal policy stance too is shifting - albeit somewhat more gradually than elsewhere - courtesy of the sovereign debt crisis, we think. While it's mainly the EMU periphery that has so far been under the market spotlight, France could be affected too at some point - should concerns spread to core countries. Indeed, the government has recently announced that, in order to achieve the goal of cutting the budget deficit to less than 3% of GDP in 2013, it will freeze nominal government spending (excluding interest payments and pensions) over the next three years. But future attempts to replenish the state coffers - which we deem to be very likely - might well entail changes in the current tax breaks and, possibly (but more difficult in the short term), the pension system.
What's the Impact on GDP Growth?
Different Country, Different Policy
Sooner or later, then, we expect all the major euro area economies to go through a period of fiscal tightening. But just what impact will that have? The effects of government policy will depend on a wide variety of factors, not least the types of policy pursued by each government. Analysis based on previous periods of fiscal consolidation, including work by the OECD and IMF, suggests that income tax hikes are particularly harmful as they discourage work. Government spending cuts are thought to do less damage because they can lead to efficiency improvements. Of course, the fact that there is scope to pursue certain policies does not mean this is what governments will do, and the likely method of consolidation remains highly uncertain. On the face of it, though, a given degree of belt-tightening might be quite damaging in Greece, Ireland and Spain, due to tax hikes. Germany and France might fare a bit better, given that income taxes are unlikely to rise in the former and spending could easily be cut - to various degrees - in the latter.
The Fiscal Multiplier - A Short Review
The actual impact on aggregate demand needs not coincide with the amount of fiscal tightening. Indeed, there is no consensus in the academic literature on the so-called fiscal multiplier. For example, the OECD Interlink model (see Dalsgaard et al., 2001) has a multiplier equal to 1.2 for government spending, i.e., a cut in government spending of 1pp of GDP reduces euro area GDP by 1.2% over one year relative to the baseline - the effect, according to this model, is stronger in Germany and weaker in France, with Italy's response in between.
Similarly, a tax hike has a multiplier equal to -0.5, i.e., a hike in income taxes of 1pp of GDP reduces euro area GDP by 0.5% over one year relative to the baseline. The more recent OECD global model (see Hervé et al., 2010), however, finds a more moderate impact on GDP from a cut in government spending, to the tune of 0.8%. The various models in use at the European central banks, including the ECB's area-wide model, suggest that a cut in government spending of 1pp of GDP typically reduces GDP by 0.8-1.2%, depending on the member country (see Fagan and Morgan, 2005).
With fiscal policy being a key economic driver, the features of the INSEE MZE-2003 model might come in quite handy, because the published documentation explicitly mentions quarterly multipliers for government spending, as well as direct and indirect taxes (see Beffy et al., 2003). The INSEE MZE-2003 model has a multiplier equal to 0.9 for government spending, i.e., a cut in government spending of 1ppt of GDP reduces euro area GDP by 0.9% over one year relative to the baseline. The multiplier for income taxes is -0.3, while that for indirect taxes is -0.7.
Read Full Article »