UBS: Recessionary Risks Are Rising

UBS Investment ResearchThe conflict between "moral hazard" considerations and the need to stabilize real estate prices via a comprehensive reduction of foreclosures poses a public policy dilemma that could delay agreement on and limit the effectiveness of public policies on foreclosures. However, in the view of UBS mortgage strategists: "a bailout of some sort is an appropriate policy response to themounting credit crisis. A slight improvement in expected losses on outstanding subprime and Alt-A deals would go a long way toward restoring stability." Downside Risks Rattling Fed 􀂄 More recession evidence The latest data, including the 63,000 decline in payrolls in February, added to theevidence that the economy is in recession, setting the stage for an easing ofinflation pressures. So far, the rate of contraction implied by the broad growth datalooks fairly modest. We are not forecasting a deep recession, due in large part toonly a small drag from inventories this time. Still, even if the recession isultimately "milder than average," as we project, data in coming months are likelyto show additional weakening. Meanwhile, ongoing financial markets problems raise the risk of greater-than-expected weakness. Against that backdrop, Fed officials showed concern by announcing new liquidity boosting measures for shortterm funding markets. We expect that concern will also be reflected in another large cut in the funds rate at the March 18 FOMC meeting: we now expect a 75 bp cut at that meeting (to 2.25%), instead of 50 bps. In addition, we now forecast an eventual lowering of the funds rate to 1.5% (by August) instead of 2.0%. 􀂄 When will it end? We believe the latest back-to-back nonfarm payroll declines are a prelude tofurther recessionary news at least through mid-year. Whether a mild recoverystarts, as we currently expect during Q3(08), hinges on materially reducing theexcess vacant inventories that are depressing home prices. That requires furtherreductions in homebuilding, a positive sales response to more affordable homes,and public policies starting to trim recently elevated foreclosures. 􀂄 The week ahead We expect the February CPI report to show tamer overall and core inflation thanthe January report a month ago. Retail sales were probably neither especiallystrong nor weak in February, consistent with consumer spending slowingsignificantly but not collapsing. More recent confidence data likely showed acontinued downtrend into early March. The rate of growth in employment-based withheld taxes hasslowed sharply, consistent with weakening in wage income andemployment growth in the last few months. At the end of last year, almost one in every 12 mortgages waseither delinquent for over 30 days or in foreclosure. More Recession Evidence The latest data added to the evidence that the economy is in recession, setting the stage for an easing of inflation pressures. The February employment report showed a second consecutive decline in total payrolls (-63,000 after -22,000) and a third straight decline in private payrolls (-101,000 after -26,000 and -14,000). More positively, the unemployment rate fell 0.1 point in February, to4.8%, although the improvement reflected a larger decline in the labor force(-450,000) than in the household survey measure of employment (-255,000). At 4.8%, the level of the unemployment rate is still up significantly from the low of 4.4% in March 2007. Meanwhile, the trend in employment and growth broadly looks more than weak enough to push unemployment up further. (We continue to forecast a 6.0% unemployment rate by year-end.) Along with the payrolls data, that is the message from the latest ISM surveys (including the composite indexes and the employment components), consumer confidence surveys (including the labor market related parts), and employment-based tax receipts. So far, the rate of contraction implied by the broad growth data looks fairlymodest. We are not forecasting a deep recession, due in large part to only asmall drag from inventory cutting this time. Still, even if the recession isultimately "milder than average," as we project, the data in coming months arelikely to show additional weakening. Meanwhile, ongoing problems in financial markets raise the risk of greater-than-expected weakness. Credit spreads have been widening sharply and equities have been falling. Against that backdrop, Fed officials showed their concern by announcing a plan to increase lending through the new Term Auction Facility (TAF) program to $100 billion from $60 billion and to add another $100 billion of liquidity toshort-term funding markets through 28-day term repurchase agreements. Theannouncement came 15 minutes before the February employment data werereleased. According to the Fed's statement, the two initiatives were intended to "address heightened liquidity pressures in term funding markets. " We expect that aggressive effort to prevent disarray in financial markets and a severe recession will also be evident at the March 18 FOMC meeting: we now expect a 75 bp cut in the fund rate at that meeting (to 2.25%), instead of 50 bps. In addition, we now forecast an eventual easing to 1.5% (by August) instead of 2.0%. After that, we expect the Fed to go on hold for almost a year before starting to push the funds rate up again in late 2009 (to 2.0% by the end of 2009). A Fed staffer emphasized that the increased funding through TAF and repooperations would largely be offset elsewhere on the Fed's balance sheet,consistent with what has been occurring already. (Otherwise, the funds ratewould plunge.) Securities held outright by the Fed will likely continue to decline sharply (see charts below). However, the TAF and repo operations will enable the Fed provide more targeted liquidity for term funding markets. Effective End to Debate Over Whether in Recession? According to the National Bureau of Economic Research's (NBER) web site:"A recession is a significant decline in economic activity spread across theeconomy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales." That is not a very precise definition, and use of the word "normally" allows discretion in interpretation. Although recessions are usually marked by at least two consecutive quarters of declining real GDP, the most important requirement has typically beensustained, outright contraction in employment. Indeed, based on current (as well as originally reported) data, real GDP did not show back-to-back declines in the 2001 recession, reflecting exceptionally strong productivity growth after the late-1990s' business investment boom. That scenario seems unlikely in the current cycle, with the trend in productivity growth more modest now than in 2001: A declining trend in employment would most likely be associated with contraction in GDP, at least eventually, after revisions. Weak and Weakening So far, total payrolls show only two months of decline and, in part, the declinecould have reflected mere volatility. Private sector payrolls have declined inthree straight months, but significantly in only one month. Still, as noted earlier, other employment as well as non-employment data are also showing clear weakening in the last couple of months, making a "turning point" highlycredible. Along with the unemployment rate, other economy-wide data showing significant weakening include the employment and composite indexes in the ISM surveys and tax receipts. Meanwhile, the financial markets backdrop provides little basis for believing the downward momentum will soon bereversed.Declining Employment The table below shows details for the payrolls data in recent months. The netdecline in payrolls in the last three months has been quite broad-based,extending to the manufacturing, construction, wholesale trade, retail trade,information, financial, and "temps" sectors. While the household surveyemployment data have been exceptionally volatile, they also show a net decline in the last three months. If anything, we suspect the weakness in payrolls is being understated; there has been a tendency in the past for payrolls gains to be overstated initially when growth is slowing significantly, with downward revisions later. That tendency can be seen in the tables below, which show the pattern around the start of the last two recessions. Estimating payrolls is especially challenging at turning points because of sharp changes in the number of jobs gained and lost from firms starting up and shutting down, respectively. BLS officials have changed their methods over time, and currently apply a model-based "net birth-death adjustment" to the data, but they acknowledge there is still plenty of scope for miscounting initially. The 450,000 decline in the labor force in February helped lower theparticipation rate to 65.9% from 66.1%. As noted, employment in the household survey also fell, albeit by "only" 255,000, and the number of unemployed fell 195,000. The labor force participation rate is cyclical, tending to weaken relativeto the secular trend during recessions (with workers effectively giving uplooking for a job amidst a weak job market), albeit not to the degree of the 0.2point February decline; a 0.2 point drop in a single month likely reflectsvolatility. The cyclicality has not prevented the unemployment rate fromtrending higher in past recessions. On a smoothed, 13-week moving average basis, the y/y growth rate in employment-based withheld taxes has slowed to around 3% from 4 1/2% in early January and 6 1/2% on average in 2007. The recent slowing appears to be much more dramatic than can be explained by weaker growth in bonus income alone. Excluding tax rate changes, the pace slowed from 8.2% y/y in 2000 to 7.6% y/y in Q1(01) and 3.4% in Q2(01). (The recession officially began in April 2001.) Until recently, initial jobless claims were showing virtually no rise. However,they have risen sharply since mid-January, with the four-week average up to360,000 from a low of 316,000 in January. Continuing claims have continued to rise. Broader Growth Data Also Show Clear Weakening Our estimated overall all-economy ISM index rose to 49.1 in February from45.6 in January, but that level was still low enough to suggest contraction inoverall real GDP (see chart). The non-manufacturing ISM index rose to 49.3 inFebruary from 44.6 in January. The manufacturing index fell to 48.3 from 50.7.The index of hours worked in the employment report declined 0.1% m/m inFebruary following a 0.4% decline in January. The index is down at a 1.7%annual rate so far in Q1 (versus the Q4 average). Consumption Likely to Decline for First Time Since 1991 Until recently, the drag from housing was still being offset by growthelsewhere--with strength in exports, solid growth in overall business fixedinvestment (led by structures), and moderate growth in consumer spending. So far, export growth still looks strong, but consumer spending and businessinvestment appear to be weakening. Real consumer was roughly unchanged in December and January but consumer confidence has continued to fall, suggesting a declining trend soon. In part, the weakening in confidence and spending likely reflects a fading boost from home-equity extraction as home prices decline and lending standards are tightened. We estimate "active" homeequity extraction, including home equity loans and cash-out refis, fell to 1.7% of disposable income in Q4 from 3.2% in Q3 and 4.0% in Q2 (see chart below). Real consumer spending last showed a quarterly decline in 1991; it continued to rise during the 2001 recession. Light at End of Tunnel? If, as seems highly likely, a downturn is now under way, then data are likely tocontinue to look recession-like for at least several months. The following article focuses on "When Will it End?" The housing sector triggered the recession and will most likely be key to the eventual turnaround. When Will It End? In our view, the back-to-back nonfarm payroll declines in the first two monthsof this year will be followed by more downbeat recession news through midyear. We continue to foresee annualized real GDP growth of -1.0% in Q1(08) and -1.5% in Q2(08) before a recovery to 1.5% in Q3(08) and 2.8% in Q4(08). However, if there's a second-half recovery that is correctly anticipated by the financial markets, the bad news for investors may not extend too much longer. (For instance, see historical lead times between troughs in the stock market and the economy in accompanying table.) Whether the recession ends during Q3(08) hinges critically on evolving conditions in the housing market--the recession's primary causal factor. Recessions importantly reflect inventory corrections, which must be well on the road to completion before an overall business downturn can end. Whetherexcess inventories represent manufactured goods, as has often been the case in past recessions, or structures, as in the current setting, the business cycle mechanics are similar. In a market economy, over-supply generates downward price pressures prompting suppliers to reduce production below demand levels. However, as excess supplies are whittled and downward price pressures abate, output recovers toward demand even if demand is at less than its earlier peak levels due to negative spread effects emanating from the output reductions. In the current setting, unusually high levels of vacant residences for sale havegenerated sharp reductions in home prices. (See accompanying chart.) Thus,housing starts must temporarily fall beneath underlying demand until excessinventories are reduced enough for currently falling home prices to stabilize. Given the unusual severity of the earlier over-production of housing units, theaccompanying corrections in housing and related production and in home prices have been enough to drag the overall economy into recession. And a recovery in the overall real GDP will not begin until this major drag on overall economic activity considerably abates.So far through early 2008, some important information has not indicated an imminent end to the housing correction. For instance, the sharp 3.0% Januarydecline in private residential construction (see accompanying chart) raises the possibility that housing's direct subtraction from quarterly real GDP growth might be even greater than the 1.25 percentage points that it subtracted from just 0.6% Q4(07) real growth. And on March 6, it was reported that in Q4(07) there were further rises in home mortgage delinquency and foreclosure rates. (See accompanying charts.) According to Fed Chair Bernanke's remarks on March 4,over half of foreclosure starts result in the sale of the mortgaged property--one of the critical factors currently depressing home prices. However, when the Fed fosters lower interest rates during a recession, theclassic query is whether the Fed is only impotently "pushing on a string." Inother words, when interest rates fall in recessions, it is frequently said that "you can lead a horse to water but cannot make it drink." In the current setting, one doubt about the efficacy of monetary policy stems from bank tightening oflending standards. Nevertheless, in the specific circumstances of the current recession, lower interest rates should be helpful for the housing industry. (1) Tighter lending standards entail reducing maximum allowed debt service burdens. For home mortgage lending, more restrictive lending standards areaccompanied by reductions in the allowable ratio of monthly debtrepayments to income. And as lower borrowing costs reduce monthly debtrepayments, more borrowers can meet the more restrictive standards.(2) For the just over 1 million subprime borrowers with adjustable rate mortgages re-setting in 2008, a lower Fed funds rate reduces incrementalreset debt burdens. (See accompanying table on next page.) This, in turn,should make delinquency and foreclosure rates somewhat lower thanotherwise. And containing foreclosures and the accompanying home salesis a key to stabilizing house prices. However, as Fed Chair Benanke discussed in his March 4 remarks titled"Reducing Preventable Mortgage Foreclosures," new public policies outside of monetary policy are necessary to address the foreclosure problem. Indeed, as discussed in the March 4, 2008 edition of the UBS Mortgage Strategist weekly report, there have been high default rates even before interest rate resets. This partly reflects the behaviour of investors and resident owners who eitherfraudulently obtained their loans or made unwise personal financial decisions. In the view of UBS mortgage strategists: "Any bailout plan that excludes most of these homeowners will almost by definition have only a marginal impact on the 'subprime housing crisis.' We believe that many authors of bailout plans fail to take this into account. In order to be effective, a plan will have to help many homeowners that most would prefer not to help." Meanwhile, amidst the debate over public policies addressing the residential real estate crisis, could there be an evolving "market solution" to the excess supply problem in residential real estate? Until recently, most observers believed that it was impractical to rely much on lower prices to clear residential real estate markets, because national average home prices historically have been sticky. However, U.S. history is replete with examples of significant downwardreal estate price adjustments where there have been depressed local real estate market conditions. And on a national basis, in Q4(07), there was a faster rate of decline in the S&P/Case Shiller home resale price index.(See accompanying chart.) In principle, as prices fall, the incremental supply (i.e., production) ofhousing declines and unit demand is aided by improved affordability. (Seeaccompanying charts.) And once excess supplies are sufficiently trimmed, home prices should start to stabilize. While the price system will play a role in helping clear the excess supply ofresidential real estate properties, new public policies will remain critical.Relying solely on the pricing mechanism to clear the residential real estatemarket runs the risk of potentially quite significant negative real estatewealth/borrowing effects on household spending. Also, as home prices decline, more households experience negative home equity, which can trigger higher loan defaults and foreclosures and related further downward pressures on home prices. On the demand side, unit home sales need to start recovering. A key wildcard is household formation. On the supply side, homebuilders' output needs to continue declining. A combination of higher sales and lower levels of housing output and foreclosures should result in reduced excess inventories. In our view, the inventory measures best correlated with prices are measures calculated with the quarterly vacancy data available a month after a quarter. These are the homeowner vacancy rate and the months' supply of unoccupied housing units. On a more timely monthly basis, the monthly new and existing home inventories statistics will continue garnering considerable investor interest. And on an even timelier basis are our proprietary weekly data on residential real estate listings.