Likewise, refinancing should accelerate the decline in household debt service in relation to income to a sustainable 11-12% range by late this year, with ongoing benefits to discretionary income and creditworthiness. We think that American consumers are deleveraging their balance sheets and rebuilding saving faster than expected. Consequently, the headwind to consumer spending from deleveraging will be a smaller risk to the outlook, as consumers now can spend more of their income.
Not a panacea for housing woes. Despite the swelling number of refinancings, the total refi windfall is still small in relative terms - it's only 0.1-0.2% of disposable income. Clearly, lower rates and debt service through refinancing alone are not a panacea for housing woes. That's partly because only about 20% of the $10.2 trillion in household mortgages outstanding can be refinanced under the current eligibility criteria. More important, we think the real key to resolving the foreclosure crisis is reducing or writing down principal, not simply lowering monthly payments.
Two major hurdles block many refis: First, many of the borrowers with positive equity aren't eligible for refinancing because they don't meet origination criteria. Those include ensuring that the loan is no more than 80% of the appraised value of the property, plus verification of the borrower's FICO score and income. Originators faced with ‘putbacks' from the GSEs (Fannie and Freddie) on prior loans are understandably skittish to refinance less-than-pristine borrowers. In addition, until now, the 25% of borrowers whose property values are underwater could not refinance. They have had to choose among four tough options: negotiating a principal write-down, a short sale, foreclosure or strategic default. Our colleagues Vishwanath Tirupattur and James Egan find that strategic defaults - in which underwater borrowers who can afford to service their mortgage nonetheless decide to mail the keys back - have risen from 12% to 18% of all defaults in the past six months.
Indeed, the experience of the various programs to modify or refinance mortgages suggests that reducing principal is essential for success. Mortgages modified under the Home Affordable Modification Program (HAMP) have 50%-plus re-default rates. And the scope of the HAMP program has shrunk. Only about 350,000 permanent HAMP modifications are in place, and more than twice that many borrowers have fallen out of the HAMP trial modification program. At this rate, HAMP will hardly reach the 3-4 million borrowers that the administration targeted.
More encouragingly, the US Department of Housing and Urban Development (HUD) in August provided details on the enhancements to its refinance program which allows lenders to provide additional refinancing options to underwater borrowers. As of September 7, certain underwater non-FHA borrowers who are current on their existing mortgage, and whose lenders agree to write off at least 10% of the unpaid principal balance of the first mortgage, were given the opportunity to qualify for a new FHA-insured mortgage. Encouraging principal write-downs or principal forgiveness will reduce the incentive for future strategic defaults or re-defaults by reducing borrowers' negative equity and providing an incentive for them to service the loan.
Potent Rx for housing ills. Three remedies could significantly improve the housing and mortgage markets. First, the Fed could force rates significantly lower by buying more Treasury and mortgage securities, which would increase the refi windfall for eligible borrowers. Second, if the federal government formally recognized its existing guarantee on many mortgages, it would expand the number of borrowers eligible to refinance by about 18 million, doubling or tripling the aggregate windfall at current rates. David Greenlaw estimates that such a scheme could save those borrowers about $46 billion annually (apart from transaction costs). Finally, expanded use of principal write-downs or forgiveness would not only reduce monthly payments, but also reduce the probability of default, strategic or otherwise. In combination, these three options would be a potent Rx for housing ills.
Is real debt service more onerous today? Not really. Some observers think that the benefits of declining debt service are overstated. They argue that the debt service burden is more onerous in today's low-inflation setting than in the high-inflation 1970s and early 1980s. Back then, high inflation eroded the real value of debt, so debt service should be adjusted to account for inflation. That point has some validity. But high inflation also eroded incomes. So it's appropriate either to look at nominal debt service relative to nominal incomes or to use real figures for both, but not a mix. On that score, real borrowing costs today are half what they were in that earlier period of high inflation. For homeowners as a class, lower real rates have reduced real debt service.
But that development does little for those borrowers stuck with high-coupon mortgages and unable to refinance. For many, plunging home prices have shrunk - and could continue to shrink - homeowners' equity. That leaves insufficient incentive to stay current on their mortgages, encouraging at best delinquency and at worst foreclosure or strategic default. For borrowers who can't refinance, today's low inflation rate has indeed increased the real cost of their stranded debt. That's why additional housing remedies are still needed.
Rising expectations for increased government support boosted Treasuries to good gains in the short and intermediate parts of the curve over the past week even as the run of improved economic data reported through September after a dismal August continued, putting some pressure on the long end. The substantial recent renewed rise in worries about the budget strains in the euro-zone periphery also continued as a background source of support. In the first part of the week, government support hopes were focused on the Fed, with the market increasingly expecting that the FOMC will signal after its meeting on Tuesday that renewed quantitative easing is imminent. In the second half of the week, expected buying from the Japanese government added to the upside, though only in the shorter end of the curve, as investors anticipated the $15-20 billion of dollar buying Japan's Ministry of Finance was estimated to have done to halt the yen's run-up being invested, with expectations that this buying would be concentrated around the 2-year area. Meanwhile, an increasing number of market participants seem to expect that the Fed will signal in Tuesday's FOMC statement that a resumption of large-scale quantitative easing is reasonably likely to be announced after the November FOMC meeting. How the Fed decides to include its deliberations about the circumstances under which it would resume easing will clearly be the key sentence in the statement, and we think some version of Fed Chairman Bernanke's Jackson Hole formulation - "the Committee is prepared to provide additional monetary accommodation through unconventional measures if it proves necessary, especially if the outlook were to deteriorate significantly" - will likely be included. The bar for resumed easing is probably lower than it was when he spoke on August 27 as expectations for sub-par growth through at least year-end have become more entrenched, and the risks of a significant fiscal tightening early next year have risen as Congress has been unable to reach agreement on extending some or all of the Bush tax cuts. But the economic outlook hasn't "deteriorated significantly" in recent weeks. Key data releases have actually been broadly improved throughout September since the Chairman spoke in late August, including the key figures released over the past week. So to the extent that investors are expecting a clear indication from the Fed that resumed QE in the near term is likely, we think that they will probably be disappointed. The run of better data that began on September 1 with the very surprising gain in the ISM continued in the latest week with a solid retail sales report that showed a good gain in ex auto sales during the key back to school shopping season and left 3Q consumption on pace for another gain near 2% and GDP running near +2.6%, small improvement in the early regional manufacturing surveys after significant weakness last month that pointed to another elevated ISM in September, and a further improvement in jobless claims that suggested a notable pick-up in underlying job growth this month after the recent run of sluggish results. Industrial production growth was sluggish but mostly because of a partial correction of the surge in auto assemblies in July, with ex autos manufacturing output posting another solid and broadly based gain.
For the week, substantial short and intermediate gains and small long-end losses sharply extended the curve-steepening reversal that has been underway since Fed Chairman Bernanke's Jackson Hole speech three weeks ago, with 2s-30s up 14bp on the week and 43bp since August 26 and 5s-30s up 18bp and 33bp. The 2-year yield fell 10bp to 0.47%, 3-year 12bp to 0.75%, 5-year 14bp to 1.44%, 7-year 10bp to 2.13% and 10-year 5bp to 2.74%, while the 30-year yield rose 4bp to 3.91%. There was some notable upward pressure on overnight financing rates that flowed through into a bit of upside in T-bill yields. The continued flood of Treasury supply has been gradually shifting Treasury overnight repo rates higher since the SFP bill program was ramped back up starting in February. With the $65 billion in net coupon supply hitting the market with the settlement of the 10-year TIPS, 3-year, 10-year and 30-year auctions on Wednesday on top of a big corporate tax day, the overnight general collateral Treasury repo rate averaged a post-QE high of 0.31% Wednesday and stayed elevated at 0.30% Thursday. It fell back to a still somewhat elevated 0.25% on average Friday but late in the day was up near 0.35%, suggesting possible renewed elevation Monday. With this backdrop, the 4-week T-bill yield rose 2bp on the week to 0.11% and 3-month 1bp to 0.15%. This flattened the 3-month/2-year spread by 11-32bp, matching the low since March 2008 previously hit on August 24.
TIPS lagged, especially in the 5-year sector, as the inflation reports for August were more restrained than expected, oil prices fell 4%, and there was sticker shock after the 10-year TIPS yield reached an all-time low close Tuesday of 0.895%, just below the prior low reached in March 2008 during the week that Bear Stearns collapsed. The 5-year yield fell 4bp on the week to 0.13%, 10-year yield fell 2bp to 0.96%, and 30-year yield rose 7bp to 1.78%. The consumer price index rose 0.3% in August, lowering the year-on-year rate a tenth to +1.1%, as energy prices surged 2.3% as a result of gasoline prices holding steady instead of seeing the normal big drop in August. The core slowed to +0.05% after gains averaging 0.14% in the prior three months, keeping the annual rate at +0.9% for a fifth month. Softness in OER (0.0%) and rents (-0.1%) after slight upside in prior months and a 1.3% drop in hotel prices left the key shelter category unchanged on the month. Widespread industry data suggest that this softness in rents and hotels will likely be only a temporary blip. Also contributing to the core CPI moderation was a flat reading for education, with tuition costs posting their smallest year-on-year rise on record. The PPI report also moderated on an underlying basis. The overall producer price index rose 0.4% in August, lowering the year-on-year rate to +3.1% from +4.2%, as a dip in food prices (-0.3%) partly offset a surge in energy (+2.2%). The core moderated to +0.1%, reducing the annual rate to +1.3% from +1.5%, after a five-month run of slightly elevated +0.2% readings excluding motor vehicles volatility. Often volatile auto prices were little changed this month, and both core consumer goods (+0.1%) and capital goods (+0.1%) prices posted only minor increases.
Relative performance by mortgages was poor, with investors continuing to worry about rising supply as refinancings accelerate and carry is hurt by dislocations in the dollar roll market. Against the good gains in the intermediate part of the Treasury curve, MBS prices ended little changed after lagging in particular on Friday. This left current coupon yields little changed near 3.55%, up from the record low of 3.2% hit at the end of August. Over this period, Treasuries have sold off much less, with the 5-year yield up 10bp, 7-year 20bp and 10-year 27bp. The good news, however, is that this has not flowed through much into mortgage rates, with Freddie Mac's survey showing only a 5bp rise in average 30-year rates to 4.37% over the past couple weeks from the record low hit at the end of August. Our mortgage strategists calculate that the primary/secondary spread reached an unusually wide 127bp, as the recent low in MBS yields has narrowed recently to near 105bp (see Janaki Rao and Zofia Koscielniak's Pass-Through Perspectives: FN 6s: Looking Cheap Even Considering Risk of Faster Prepays, September 17, 2010). They believe this narrowing points to easing originator capacity constraints and rising industry competition, which could help mortgage rates more closely follow a renewed decline in MBS yields going forward. Still near-record low mortgage rates should drive a continued elevated level of mortgage refinancing after the sharper-than-expected acceleration in MBS prepays in August, which should help accelerate the move in the consumer debt service ratio down to what we could consider a sustainable 11-12% level by late this year from the record high 14% hit in early 2008 (see Richard Berner's US Economics: Refis Will Hasten the Debt Service Decline, September 17, 2010). The supply pressures from rising refinancing activity has been a problem for the MBS market recently, however, and we think that the Fed should consider altering its SOMA reinvestment policies to add MBS buying to its ongoing Treasury purchases to help keep this process on track.
The most important economic data release in the past week was the retail sales report, which showed good underlying upside in August, the second most important time of the year for mall-type retailers after Christmas. Overall retail sales rose 0.4%, with auto sales falling 0.7% after the small decline reported in unit sales, but ex auto sales jumping 0.6%, the best gain since March. The biggest contributor to the upside was the heavily weighted grocery store category (+1.3%), which rebounded after five straight declines. In line with the solid chain store sales reports, which may have been helped by expanded state sales tax holidays during the key back-to-school shopping period, the clothing (+1.2%), general merchandise (+0.4%), and sports, books, and music (+0.9%) categories also all showed good gains. And higher prices boosted sales at gas stations (+1.9%) again. With the key retail control grouping up 0.6%, we continue to see 3Q consumption rising 2% for a third-straight quarter. This left our 3Q GDP forecast at +2.6%, up from a low of +1.8% at the beginning of the month.
Early indications for the initial run of key September data that will be released in early October were positive. Headline results were mixed, but on an underlying basis the Empire State and Philly Fed manufacturing surveys both showed minor improvement after significant weakening last month. On an ISM-comparable weighted average basis, Empire rose 0.6 point to 50.9 in September after falling 2 points in August, while the Philly Fed rose 0.3 point to 46.3 after plunging four points. Our preliminary forecast is for the national ISM to dip to a still quite robust 55.0 in September after the surprising to 56.3 in August. Meanwhile, initial jobless claims dipped 3,000 in the week of September 11 and 28,000 the past two weeks to 450,000, a low since early May aside from a week in July that was distorted by much lower-than-normal short-term auto layoffs. The survey week for the September employment report will be covered in next week's initial claims figures (with the late survey week shifting the employment report release to the second Friday in October on the 8th), and if the recent improvement is sustained, we will likely be looking for a significant acceleration in October job growth from the sluggish ex census gains averaging only 55,000 from May to August.
Focus in the coming week will be squarely on the FOMC meeting Tuesday, but there are a few notable economic releases, mostly housing-related, through the week that we expect overall will extend the better tone to the incoming data seen all this month. Economic data releases due out include housing starts Tuesday, existing home sales and leading indicators Thursday, and durable goods and new home sales Friday:
* We look for August housing starts to fall to a 530,000 unit annual rate. The employment report showed a rise in construction employment during August. However, the upside appeared to be concentrated in the non-residential sector. Thus, we look for the post-tax break pullback in new home construction to continue in August. Our estimate implies a 3% dip relative to July.
* We look for existing home sales to rise to a 4.20 million unit annual rate. The pending home sales index registered a slight uptick in August, and regional results that have already been released point to some upside in resales during August. So, we look for a partial 10% rebound from the sharp 27% plunge seen in July.
* The index of leading economic indicators is likely expected to post another slight gain of 0.1% in August, with positive contributions from the yield curve, money supply and workweek. Partly offsetting negatives are likely from the jobless claims and supplier deliveries components.
* We forecast a 0.8% decline in August durable goods orders. The theme of last month's durable goods report was a decent headline result but very weak underlying details. This month's report should be the opposite. We look for an outright decline in headline orders attributable to a sharp pullback in the volatile aircraft category as signaled by company reports. Meanwhile, the core category - non-defense capital goods excluding aircraft - is expected to show a solid gain (+2.5%), reflecting the bizarre seasonal pattern that has evident for quite some time now. Bookings for a wide range of items - especially machinery and electrical equipment - have been plummeting in the first month of the calendar quarter and then recovering all of the lost ground during the remainder of the quarter.
* We forecast August new home sales of 275,000 units annualized. The homebuilder sentiment survey held relatively steady at a depressed level in August. So, we look for little change in sales of newly constructed residences relative to the 276,000 pace seen in July.
Summary and Conclusions
In this report, we address the question of whether heightened political volatility can undermine our thesis that Italy is transitioning into core Europe (see Inching into the Core, March 15, 2010). We conclude that this is unlikely:
• Political events move Italian financial markets significantly. In the weeks before and after a government collapse, interest rates tend to rise and equity markets fall. And asset prices decline during the time span between the end of the previous and the start of the new government.
• Yet Italy faces weaker fiscal headwinds than most other euro area countries, thanks to the introduction of fewer stimulus measures to cushion the economic fallout of the financial crisis. So, we believe that there is no need for Italy to adopt a Greek-style austerity programme.
• Complying with the European Commission's demands is within reach. The recent bond yield spread (re)widening is not related to Italy's fundamentals, in our opinion. Conversely, it has to do with a predictable shift towards ‘risk-free assets' - triggered by low risk appetite.
• When it comes to refinancing risk, Italy's government debt features some desirable aspects, ranging from long average maturity to solid domestic support, as well as a smaller proportion of government debt coming due over the next 12 months and a not so big money market.
• Still, investors may again question Italy's long-term fiscal sustainability. Admittedly, some deficiencies are quite evident, such as poor demographic prospects. But we think that Italy has already taken several steps to deal with them, unlike several other EMU countries.
• There is a difference between ageing and the cost of ageing. Pension reform and belt-tightening have improved Italy's fiscal position. The biggest risks have more to with the interaction between the growth and fiscal prospects, rather than the public finances taken in isolation.
• Risks: on the upside, households and firms won't need to deleverage aggressively - as their balance sheets are in better shape and bank lending is less impaired than in the EMU periphery; on the downside, with an appreciable competitiveness gap relative to core Europe, growth might turn out to be too weak.
Political Risk Analysis Back in Vogue
In the face of the sovereign debt crisis, a country's ability to push through tough measures is essential. With the new parliamentary group Future and Freedom (FLI) entering Italy's political scene, the ruling coalition's members are now reduced to 297 from 330 (out of a total of 630) - not enough votes to command an absolute majority in the Chamber of Deputies. So, the confidence vote on September 28 is an important risk event for markets, although Italy is not in a unique situation - as Spain, Portugal, the Netherlands and Belgium too (accounting for 40% of euro area GDP) have minority or caretaker governments. Italy has gained credibility over the years. But this still unresolved political situation could generate additional volatility in the market and make investors more doubtful about the pace or even the direction of Italy's policies, in our view.
Do Political Events Move the Market?
Government collapses and other political shocks, as well as ‘external events' such as natural catastrophes and terrorist attacks, have exerted a substantial influence on Italian financial markets over the past 40 years, as suggested by various econometric analyses. Fratzscher and Stracca (2009), for example, find a statistically and economically meaningful effect across various asset classes.
In particular, each event in these categories has raised Italian short-term interest rates by about 8-9bp and lowered equity returns by between 0.4% and 0.9% during the period under examination (1973-2007). And nominal effective exchange rates have responded to such events too. By contrast, it does not seem that referenda and the formation of governments have exerted a systematic influence on Italian asset prices, perhaps because these events tend to be anticipated or expected.
Although the magnitude of these contemporaneous effects, taken individually, might seem quite small - i.e., neither a single 1% drop in equity prices nor an isolated 10bp rise in interest rates - we think that the cumulative effects are meaningful and quite substantial. Indeed, according to the above-mentioned empirical work, the total impact of the 131 events in the database has raised Italy's short-term interest rates by 550bp and lowered Italy's equity markets by more than 40%.
Political stability or credibility could be interpreted as some form of capital stock, which various adverse shocks depreciate little by little. The impact of each individual shock may be small, but their overall effect is a significant depreciation of the capital stock. The results of this statistical approach should be taken with a pinch of salt - not at face value. While the model assumes that the effects of political news on asset prices are permanent, in the real world some form of slow mean-reversion seems plausible.
What's more, these events may have an impact on financial markets beyond the day when they occur. And, to the extent that they are anticipated, asset prices may react even before the events themselves take place. Indeed, while some events are truly unexpected, e.g., natural disasters and accidents, terrorist attacks, and also some resignations and other political events, government collapses and election outcomes may partly be anticipated. Thus, financial markets might respond to the two groups of events in two different ways.
Take Italian government changes, for example. The time horizon under examination (1973-2007) has witnessed 31 collapses. So, on average, governments replaced one another almost once a year, though the frequency of changes has declined substantially over the past decade or so. Furthermore, the frequent turnovers were associated with a period without a formal government for an average of 35 days. This means that, over the 35 years included in the above-mentioned econometric study, Italy has been without a formal government for more than three years - i.e., close to 10% of the time.
How Do Financial Markets React Over Time?
Historically, the political risk premium on Italian financial markets has been substantial, though it might have diminished with EMU membership and more political stability in recent years. In the two weeks before and after a government collapse, the cumulative rise in interest rates is about 24bp. Similarly, equity markets fall by around 5% over the same period. What's more, asset prices continue to decline during the gap between the end of the previous and the start of the new government.
Further, there seems to be a positive reaction both before and after the inauguration of a new government. In particular, short-term interest rates drop by around 30bp and equity markets rise by 3% or so in the two weeks before and after the start of a new government. The magnitude of these asset price movements is substantial. The peculiar pattern of Italy's financial markets in response to political events is perhaps due to the frequency of government changes and their partial predictability (when approaching the event).
Conversely, repeating the same exercise for other types of events, the picture that emerges is quite different. We looked at the behaviour of asset prices around natural disasters and accidents, terrorist attacks, etc. Unlike with government changes, financial markets react immediately after the event - but neither directly before or with a two-week delay after the event. This suggests that these types of events truly are unexpected and have a permanent impact.
There is at least one major caveat. In particular, the reaction of financial markets to government changes varies substantially, not only because some outcomes may have been anticipated better than others, but because measuring whether markets considered a particular change more favourable or desirable than others is a highly subjective exercise. However, a similar analysis, applied to a broader set of countries, seems to confirm these findings, both directionally and in terms of magnitudes (see Fratzscher and Stracca, 2008).
What's more, this econometric approach - which belongs to the broad category of statistical exercises known as ‘event studies' - focuses on the very short term, i.e., days and weeks before and after the event, by definition. This means that market timing becomes more crucial than ever for investors. From a long-term perspective, however, the fundamental trends that shaped the global, European and Italian economies over the years and decades are far more important.
Indeed, these driving forces have exerted a discernible impact on Italy's markets, ranging from the oil shocks in the 1970s to the deep recession and exit from the Exchange Rate Mechanism in the early 1990s. More recently, Italy's EMU membership has contributed to a notable stability in the bond market for a whole decade, while the effects of the tech bubble or the various stages of the financial crisis of the past couple of years are evident in the ‘rollercoaster' of the stock market.
Is the Fiscal Consolidation Process at Risk?
We doubt that the possibility of a (temporary) turbulent phase in Italy's politics will eventually result in a fiscal U-turn. Indeed, with markets focusing increasingly on whether all euro area countries are on track with the implementation of their belt-tightening plans, the chances are that any fiscal slippage will be punished.
We believe that Italy's policymakers are well aware of the need to stick to the agreed austerity programme and show a firm commitment to make up for any eventual fiscal shortfall - which is the best way to convince investors, rating agencies and the various international organisations, we think. Indeed, one constant feature of Italy's fiscal policy of the past three-and-a-half years or perhaps even going back to the past decade - i.e., starting before the economic fallout of the financial crisis - has been its prudent approach to managing the public purse.
Italy faces weaker fiscal headwinds than most other euro area countries, in our view. True, the overall budget deficit is above the European Union's Stability and Growth Pact (S&GP) ceiling of 3% of GDP in Italy as in virtually all other EMU members. However, at a little over 5% of GDP last year, it is a far cry from the majority of its neighbours - courtesy of the introduction of fewer stimulus measures despite a more severe recession than in the ‘typical' euro area country. The difference with the peripheral countries, where the deficit exceeded 10% of GDP in Greece, Spain and Ireland and was close to that figure in Portugal, could not be starker.
One way to get a sense of the magnitude of the stimulus is by looking at the cyclically-adjusted (or structural) budget balance. This indicator measures discretionary changes in fiscal policy abstracting from the cycle. For example, if the economy enters recession, the budget deficit will widen almost automatically because of the workings of factors such as unemployment benefits - not because of any stimulus to the economy. Thus, the cyclically-adjusted budget balance accounts for the ‘true' fiscal stimulus. Italy's overall and structural budget deficits have risen much less than those of its peers.
Put differently, we believe that there is no need for Italy to adopt a Greek-style austerity programme - which would depress economic activity - to bring its budget deficit back below the S&GP ceiling. This is not to say that Italy (where government debt has climbed from slightly above 50% of GDP in 1970 to almost 116% last year) does not need to go through a period of fiscal austerity. Rather, it is an observation that, for the next few years, Italy's relatively small budget deficit means that the government only needs to lower its primary structural deficit by a minimum of 0.5% of GDP per year. Not only is this a much more moderate adjustment than the European Commission is recommending for Greece, Spain, Ireland and Portugal, but it is the lowest required tightening in the region together with Germany.
Read Full Article »