Like the other major economies, the euro area is now expanding again. But, more than anywhere else, the strength of the recovery hangs crucially on the behaviour of bank lending. In this report we therefore take a close look at the prospects for bank lending and at the implications for the economic outlook. We conclude that the risks of a full-scale credit crunch have diminished. However, the creditless recovery that we envisage will be slow and shallow.
What Is a Credit Crunch?
There is no universal definition of a credit crunch. For us, a credit crunch is a reduction in the availability of credit that is abnormally large for a given stage of the business cycle and the profitability of investment projects - due to balance-sheet constraints of the banking sector. In these circumstances, credit becomes less available regardless of the price that borrowers are willing to pay.
A year and a half ago, everyone (ourselves included) talked about the credit crunch as a serious risk to the economic outlook (see EuroTower Insights - Cracking the Credit Puzzle, March 4, 2008). However, that should have been more accurately labelled as a liquidity crunch for banks and non-financial corporations in funding markets. Since then - courtesy of various government measures - credit spreads have narrowed, and corporate debt and equity issuances have surged. What's more, effective interest rates on loans have been falling for a while and, lately, euro area banks have projected looser credit standards.
These developments are reassuring. However, it is too early to sound the ‘all clear', in our view. When the recovery in corporate spending will start gaining traction, firms might initially be able to draw on internal cash flows but, eventually, they will likely need external funds to boost investment and replenish their stock of inventories. The risk is that if euro area firms cannot obtain these funds, the rebound in activity following a turnaround in the inventory cycle could quickly reverse.
The yet-to-be-crystallised write-downs, timid recapitalisation and excessive reliance on ECB funding might backfire in a financial system that is still largely bank-based. Monetarists would be concerned (see Milton Friedman and a W-Shaped Recession, December 17, 2009). A credit crunch would hit corporate spending hard in the near term. In addition, more prolonged feedback effects between bank profitability and economic growth could make the credit crunch a constant feature, weighing on euro area growth in the coming years.
In contrast to the earlier liquidity crunch, there would be little that the ECB could do: the bank cannot boost banks' equity capital buffers; this would need to come from either governments or private investors. A late-cycle credit crunch could seriously curtail access to bank lending, causing the economic recovery to falter.
How Important Is Bank Lending in the Euro Area?
But how much does bank lending matter? It turns out that it matters a great deal. Indeed, the economic cycle is closely intertwined with the credit cycle in the euro area. Over the past 15 years, for example, fluctuations in bank lending to the non-financial private sector have been closely correlated with those in the broader economy, although establishing whether economic developments drive credit developments, rather than vice versa, is not straightforward.
Indeed, it is difficult to identify - at least visually - leads and lags. Although the recent downturn in GDP and bank lending was almost contemporaneous, the deceleration in GDP in the late 1990s preceded that of bank lending by about one year. What's more, while the most recent GDP data show a slight pick-up in growth, aggregate bank lending has never contracted, on a year earlier, at the current pace.
The reason for the close relationship between economic and credit growth is that the euro area relies on bank financing far more than market-based financing. Indeed, as of 3Q09, the outstanding amount of securities other than shares issued by euro area non-financial corporations stood at €780 billion, while the outstanding amount of bank loans to this sector was €4,740 billion.
In other words, of the entire debt financing of the non-financial corporate sector in the euro area, just 14.1% comes from credit markets. And, accounting for equity financing too, credit markets represent less than 9% of total financing, defined as the sum of bank loans, debt securities issued and quoted shares. To put it differently, while market-based financing is not a negligible source of financing in the euro area, bank lending is more important, accounting for 53.8% of the total.
By contrast, the US non-financial corporate sector relies more heavily on market-based financing: at the end of last year the outstanding amount of corporate bonds issued by US firms was six times bigger than that of bank loans. Meanwhile, households, lacking direct access to markets, are of course 100% bank debt-financed, both in the euro area and in the US.
What Has Already Happened to Bank Lending?
So the bank lending channel is by far the most important financing channel in the euro area. Unsurprisingly, given the severity of the recent recession, annual private sector lending growth has been slowing considerably for quite some time. But the situation has deteriorated even further over the past few months: the stock of total lending to the private sector started to shrink, pushing the annual growth rate into negative territory in September-November 2009. What's more, the breakdown of bank lending by sector shows weakness across the board: the growth rate of both mortgage loans and consumer credit has now turned negative, and growth in corporate loans is plummeting precipitously.
Although this is not encouraging, the annual rate of growth of the outstanding amounts was largely expected to enter negative territory - even under a very benign scenario. For example, assuming a stabilisation in the flow of credit at the levels seen before last summer, annual lending growth to the private sector would have turned negative anyway - courtesy of strong credit numbers a year earlier.
Demand-Side versus Supply-Side Factors
So, at least in part, the slowdown in bank lending is the result of a so-called statistical base effect - which will drop out of the annual rate of growth in 2H10. But, more fundamentally, what are the drivers of bank lending in the euro area and will its recent trend continue? The key issue here is the distinction between supply-side and demand-side factors.
Demand-Side Factors
It is natural to wonder whether the growth rate of household or corporate loans is correlated with that of consumer or corporate spending, respectively. After all, even if leads and lags between fluctuations in bank lending to the non-financial private sector and those in the broader economy are difficult to identify, there may be a stronger link at a more disaggregated level. Indeed, the correlation between household loans and consumer spending is quite strong (87%), although in this case too it is difficult to establish a causal link. The correlation between corporate lending and corporate spending, however, is weaker (37%) - perhaps because some corporate lending is aimed at financing acquisitions, with primarily financial implications - even if only corporate spending on machinery and equipment is considered.
Housing market and survey-based information unveil other features of credit demand in the euro area:
1. The housing market is an important demand-side factor. Indeed, the slowdown in residential house prices in the euro area over the past three years - still underway - is at least in part responsible for the slowdown in mortgage lending. More specifically, after the housing market peaked in late 2005, growth in mortgage loans started to decelerate too - albeit with a lag of about one year. This suggests, tentatively, that the demand for housing is an important driver of the demand for credit in the euro area - at least over the past decade or so.
2. Similarly, surveys of household intentions to purchase a car and purchase or renovate a home suggest that demand-side factors have played a role behind the slowdown in mortgage loans and consumer credit. Admittedly, the European Commission's survey suggests that household intentions to buy or build a home picked up somewhat recently. However, they remain at very depressed levels. With unemployment set to rise further, and car incentives having already expired in many countries, the chances are that households will remain reluctant to add to existing debt levels - at least in the short term.
3. Surveys on capacity utilisation depict, from a different angle, a picture analogous to that painted by surveys on household intentions. The European Commission's measures of capacity utilisation - still close to its record low - and expected production constraints point to huge spare capacity in the economy. Again, this indicates a fragile situation. No wonder that demand for corporate loans is still subdued, except perhaps for debt restructuring.
The fact that demand-side factors are partly responsible for the slowdown in bank lending is hardly surprising: all else being equal, it is quite obvious that demand for bank loans decelerates during recessions. But, naturally, the negative impact of demand-side factors on bank lending tends to wane once the economy starts recovering.
Supply-Side Factors
We think that the current slowdown in credit growth is the result of supply-side factors too. The question here is how the constraints on the supply of credit will develop over the next few quarters and to what extent they will affect lending to the broader economy.
1. The ECB's bank lending survey indicates that the supply of credit is currently constrained. This survey shows that credit standards tightened further - albeit marginally - in 3Q09. For example, the percentage of banks reporting tighter credit standards for loans to non-financial corporations was 8%. This means that, starting from credit conditions being already restrictive, banks turned even more reluctant to lend to firms.
Encouragingly, the percentage of banks tightening credit standards was considerably smaller than in previous quarters, i.e., the incremental tightening has now diminished for the third quarter in a row (although this still represents a further net tightening). What's more, banks expect to ease credit conditions, albeit very slightly, in 4Q09.
Meanwhile, credit standards for loans to households, be it for mortgages or consumer credit, have become more stringent too in 3Q09. For example, the net percentage of banks reporting a tightening was 14% for mortgage loans and 13% for consumer credit. Furthermore, banks expect to tighten credit standards further in 4Q09, though less so. In other words, the degree of tightening is becoming moderate, rather than drastic, for household lending.
Similarly, another ECB survey (September 2009) shows that the proportion of SMEs reporting a deterioration in the availability of financing was larger than the proportion of SMEs reporting an improvement (net percentage = 33%).
2. Of course, the information contained in the ECB's bank lending survey is qualitative. However, one quantitative piece of evidence comes from the recent dynamics of firms' credit lines with banks. Until recently, firms may have been drawing on their existing credit lines with banks, perhaps fearing that these facilities may not be available in the future. Because banks are powerless to stop this from happening - as they have already agreed on these facilities - the availability of credit to firms may have benefited.
However, if banks are in fact attempting to reduce their balance sheets, this is likely to be accompanied by sharp declines in newly agreed credit lines. Initially, the impact of the drying-up of this form of credit may be undetectable, because not all existing credit lines expire simultaneously. When this process reaches ‘critical mass', however, the slowdown in credit growth accelerates substantially. We have now reached that point, we think.
To Sum Up...
So, how much of the slowdown in bank lending is due to supply-side factors and how much to demand-side factors? Disentangling one set of factors from the other is very difficult and, because of limited data availability, the evidence is rather tentative anyway.
Bearing these caveats in mind, our exhibit confirms the intuitive directional relationship between the percentage of banks tightening credit standards and private sector lending growth: when credit standards tighten, private sector lending slows, and vice versa.
This evidence is far from conclusive, but it tentatively suggests that private sector lending did not slow more than the tightening in credit standards would have suggested, pointing to a more dominant role of supply-side factors behind the slowdown in lending.
However, we would not be too quick to dismiss the impact of demand-side factors. Both sets of factors are likely to have mattered a great deal. After all, the ECB's bank lending survey reported that banks were facing weak demand for loans. Moreover, the European Commission's survey continues to show that the main factor limiting production is weak demand, not financial constraints.
How Will the Supply-Side Constraints Develop?
We have so far identified various demand-side and supply-side constraints to bank lending. How will bank lending develop in the coming quarters? This depends crucially on the evolution of these various constraints going forward. As we argued previously, once the economy starts expanding again, the demand for credit will eventually start recovering.
However, the outlook for the supply of credit is more uncertain and depends on two factors: the prospects for euro area banks and in particular whether they will still incur further losses; and their funding conditions, both through the market and other channels - such as the ECB. Let's examine each of these factors in turn.
Bank Losses - More to Come?
The first factor that will affect bank lending in the coming quarters is the outlook for the banking sector itself. Commercial banks are not free to lend at will, because they need to maintain some leverage ratio between assets and liabilities. To put it differently, banks need to back their lending activities with a ‘cushion' of equity or capital. If the capital is eroded because of unexpected losses, their leverage would increase. To reduce it, they would need to cut lending or raise fresh capital. What's more, the re-regulation of the banking system further complicates the picture by adding institutional constraints that affect the economics of lending (see Credit Strategy - The Other Side of Bank Deleveraging, October 23, 2009).
Euro area banks are likely to face further losses. For example, in October 2009 the IMF estimated that euro area banks had incurred about 43% of the losses they were expected to face, while their US counterparts were at 60%. A more recent estimate by the ECB - just for euro area banks - was more optimistic, suggesting that euro area banks had incurred about two-thirds of their potential losses.
Of course, these estimates are uncertain. For example, let's assume that euro area banks incur further losses of just €60 billion (about one-third of the above-mentioned ECB's estimate). This would imply a capital shortfall of the same magnitude and require a contraction in risk-weighted assets of around €860 billion (or 9.3% of GDP) in order to keep the Tier 1 capital ratio - a key measure of leverage - unchanged at 7.3%. Assuming no reduction in risk-weighted assets, the Tier 1 capital ratio would drop to 4.9%.
We illustrate what could happen under various different scenarios. For example, should banks incur further losses of €120 billion this year, their risk-weighted assets would need to shrink by €1,740 billion (or 18.8% of GDP) in order to keep the Tier 1 capital ratio unchanged. Should banks want to increase their Tier 1 capital ratio to 10%, the required contraction in risk-weighted assets would be around €2,000 billion (or 21.3% of GDP). The bottom line is that the contraction in bank lending is likely to be significant even under relatively mild assumptions.
Easing Funding Problems
So, the losses that euro area banks still have to face will likely weigh on their ability and perhaps willingness to lend. However, another important factor behind the outlook for bank lending relates to banks' ability to secure sufficient funding. For example, because of the malfunctioning of the interbank market in late 2008 and early 2009, banks were able to fund themselves mainly through customers' deposits - not through markets. This limited the scope for making loans to firms and households.
The evidence here is encouraging. The interbank market has now normalised considerably, especially at shorter maturities. For example, the difference between 3-month interbank rates and T-bills is now back to pre-crisis levels pretty much everywhere. This means that, unlike last year, banks can now secure funding in the interbank market at a reasonable price. So, with more access to cheap liquidity, they could start lending more freely.
Indeed, the ECB's bank lending survey indicates that banks no longer see capital, liquidity and access to market funding as factors negatively affecting their lending activities. Of course, there is no guarantee that euro area banks will lend to the broader economy. After all, the ECB has provided unlimited liquidity to the banking system for quite some time.
Indeed, banks may remain reluctant to lend because of the uncertainty surrounding the economic outlook. And, with the opportunity to borrow one-year funds from the central bank no longer on the table - and just one more chance to borrow six-month funds - the risk is that banks might hoard cash even if market funding problems have now eased.
Risks of Full-Scale Credit Crunch Averted...
So, where does all this leave us? We think that the outlook for bank lending is somewhat more favourable on the demand side, and will likely improve further - although temporary setbacks are possible. The euro area recovery has now started and we expect it to continue at roughly the same - admittedly rather pedestrian - pace.
Indeed, various surveys on spending intentions and capacity utilisation have shown some tentative sign of stabilisation. What's more, structural weakness in the housing market is likely to persist primarily in Spain and Ireland - not in the major euro area countries.
However, the outlook for the supply-side drivers of bank lending is more uncertain, for two reasons:
• First, euro area banks still have to incur further losses; it has taken almost two years for them to incur about two-thirds of their total potential losses - at best - and it is reasonable to assume that the remaining losses will also take place over a prolonged period. Clearly, this might exert a negative effect on bank lending.
• Second, now banks still have access to plenty of liquidity, through both markets and - less so - the ECB. There is still no evidence that this liquidity is feeding through to the broader economy, but at some point banks might feel more encouraged to lend - given the slight improvement in economic prospects.
How long will the constraints to the supply of credit remain in place for? Coming up with a precise timing is impossible, but - given the outlook for the banking sector and the reluctance of banks to lend - we believe that they will last throughout this year, although they will progressively wane.
...but Creditless Recoveries Are Sluggish Affairs
Although risks of a full-scale credit crunch have diminished, it is perhaps too early to sound the ‘all clear'. The chances are that the robust credit expansion witnessed over the past decade, during which private sector lending rose by more than 50 percentage points to 153% of GDP, will not be there to support the euro area economy when the ‘green shoots' of recovery will finally flourish.
How long this period of constrained credit availability will last for is of course an open question, but one thing is for sure: creditless recoveries are sluggish affairs. Indeed, a report by IMF economists (see Claessens et al., 2008) suggests that credit crunches tend to precede economic recessions by about three quarters. And this is the case for other financial downturns too, such as equity (five quarters) or house price busts (three quarters). Perhaps more importantly, an economic recovery is possible even before credit growth returns to positive territory: economic recessions typically end two quarters before credit crunches end and nine quarters before house prices bottom out, while equity prices tend to bottom out just as economic recessions end.
To put it differently, financial downturns tend to last longer than economic recessions. In particular, episodes of credit crunches generally last twice as long as recessions. What's more, during these episodes the real economy typically starts recovering while credit is still contracting.
While this may sound encouraging at first sight, the bad news is that these creditless recoveries tend to be much weaker than recoveries in which no credit crunch has taken place. During these creditless recoveries it generally takes much longer for GDP to return to its pre-recession level.
True, firms and households could get financing from sources other than banks, or switch to less credit-intensive activities. But that can only happen over time. As long as the bank lending channel remains impaired, it is difficult to foresee the euro area economy going back to a sustained growth rate in the near term.
How Weak Will the Creditless Recovery Be?
So, although we are well aware that any quantitative assessment of this kind is surrounded by huge uncertainty, how far will lending fall short? And to what extent will this hold back economic growth?
For a start, to be consistent with trend growth in the real economy of around 2.0% and consumer price inflation of around 2.0% (the rate which over the past 15 years or so has been consistent with the GDP deflator growing at 1.8%), nominal lending needs to rise by between 7% and 8% each year. Indeed, in order to generate the 3.9% average nominal GDP growth seen since 1995, bank lending to the non-financial private sector has had to grow by an average of 7.7%. Annual lending growth of, say, 7% is equivalent to an annual flow of net new lending of around €630 billion. This compares with an estimated drop in lending of €60 billion in 2009.
How much will this constrain economic growth? Since bank lending has never contracted in the euro area, we cannot draw any parallel from previous episodes. However, we can assume that the historical relationship between GDP growth and private sector lending growth over the past 15 years continues to hold. Over this period, an increase of one percentage point in private sector lending has roughly been associated with an increase in GDP of around half a percentage point. This means that our GDP growth forecast of 1.2% in 2010 is consistent with a flow of net new lending of approximately €65 billion.
There are two caveats. First, the past 15 years coincided with a period of financial deepening and leveraging that might not be repeated over, say, the next decade. In other words, the relationship between GDP growth and private sector lending growth might now be different. Second, there is a bit of circularity in this argument. What we are interested in, after all, is not the slowdown in credit because of demand-side factors, but that due to supply-side factors. When the economy starts recovering, this will translate into stronger demand for bank loans. The question is whether the supply-side constraints will remain in place and prevent a swift economic recovery.
Hence, we built an econometric model that takes into account the simultaneous evolution of GDP, loans to the non-financial private sector, inflation and the lending rate. The model does not include explicitly the various demand-side and supply-side factors, but has the advantage of accounting for the feedback and lagged effects among the above-mentioned variables without imposing any restriction; in this model, the economic cycle affects the credit cycle, and vice versa.
Our simulations point to weak credit growth even in our optimistic scenario, which assumes that the economy starts to expand at its pre-crisis trend rate of around 2-2.5% (annualised) from end-2009 (see Appendix for details). Under our base case scenario of a sub-par recovery this year and next (we expect the euro area to expand by 1.2% in both years), credit growth is likely to turn positive no earlier than end-2010. What's more, our pessimistic scenario of economic stagnation throughout the forecast horizon (end-2011) is consistent with a prolonged and damaging credit contraction.
The main takeaway is that it would take a sustained and above-trend recovery to trigger a strong acceleration in bank lending. The chances are that hopes of that happening - at least over the next couple of years - will be disappointed.
Implications for the ECB
We think that the risk of a full-scale credit crunch has probably been averted, but credit growth will likely remain negative for most of this year even under benign (and, admittedly, rather unrealistic) assumptions, and pick up very gradually thereafter. We believe that the ECB is in no rush to hike interest rates. We expect the first move, to the tune of 25bp, no earlier than 3Q10 - followed by a move of the same magnitude in the following quarter.
Risks to our call are slightly asymmetric. At this stage, the high uncertainty over the economic and credit outlook suggests that the ECB might remain on hold for the whole year. Should one of the key surprises that we indentified for 2010 materialise, namely a late-cycle credit crunch, this could potentially pave the way for a more dovish monetary policy stance relative to our forecast (see Europe Economics - Transition Towards a Tepid Recovery, January 4, 2010).
In order to trigger such a change in the monetary policy stance, we would need to see a deceleration of credit growth that goes considerably beyond the slowdown to be expected on the back of lower demand. For now, we are more inclined to think that the euro area is heading towards a creditless recovery. Hence, official interest rates are likely to stay below ‘neutral' in 2011 too. We expect the ECB to continue to hike gradually; we see official ECB rates at 2.50% at the end of next year.
Conclusions
In all, we see four main economic takeaways:
• First, the most intense phase of disruption in the flow of credit is probably behind us and risks of a full-scale credit crunch have diminished. In particular, the outlook for the demand-side drivers of bank lending has improved: surveys depict a rosier picture and the economy is now expanding.
• Second, the outlook for the supply-side drivers of bank lending is more uncertain. One positive factor is that banks have plenty of opportunities to fund themselves through both markets and - less so - the central bank. One negative factor is the high uncertainty over potential bank losses.
• Third, a revival in economic activity is possible even ahead of a revival in bank lending - and is indeed happening. However, in the creditless recovery that we envisage the pick-up in GDP growth is likely to be slow and shallow. We expect a sub-trend rate of expansion (a touch above 1%) in both 2010 and 2011.
• Fourth, with inflation pressures likely to remain subdued this year and next, the ECB will be in no rush to hike interest rates. We expect the first 25bp rate hike in 3Q10, followed by another hike of the same magnitude in the following quarter. We see official ECB rates at 2.50% at the end of next year.
Appendix - Modelling Euro Area Loans to the Private Sector
Our econometric approach is similar to that used by Calza et al. (2001 and 2003). It is based on the joint examination of the historical behaviour of loans to the non-financial private sector, GDP and the lending rate - all in real terms.
In this stylised model, GDP represents the demand-side factor, the lending rate is the supply-side factor. We used a multi-equation error-correction model implemented in two stages (see, for example, Johansen, 1988):
• First, we estimated the parameters of the model taking into account both the long-run economic relationship among the variables and their short-run statistical properties. We used a multi-equation framework because the distinction between the dependent variable and the explanatory variables is rarely clear-cut.
• Second, we simulated various scenarios for loan growth until end-2011. In particular, we considered three scenarios: 1) GDP follows our forecast; 2) GDP is kept constant; and 3) GDP expands at trend - i.e., approximately twice as fast as in our forecast. Inflation is adjusted accordingly. The lending rate increases by 25bp per quarter in all the scenarios.
In a nutshell, the error-correction mechanism is such that loans to the non-financial private sector will grow faster (slower) whenever their level in the previous quarter is below (above) the theoretical equilibrium level - or ‘fair value' - given by the long-run relationship. The small magnitude of the error-correction term indicates that deviations from the long-run relationship are absorbed very gradually. The model explains about two-thirds of the changes in loans to the non-financial private sector. Other variables - such as the surveys on banks' credit standards - might have increased the model's explanatory power, but their short history prevented their inclusion in the regressions.
In the long run, GDP is the main driver of euro area loans. The magnitude of the elasticity is in line with that of other studies: if GDP increases by 1%, euro area loans increase by 1.85%. The magnitude of the (semi) elasticity of the lending rate is remarkably small. Taken at face value, this suggests that - in the long run - the cost of borrowing plays a minor role. One explanation could be that, within a ‘normal' range, variations in the lending rate encourage/discourage households and firms to borrow only at the margin. At extremely high or low levels, however, the lending rate might become more important. The model does not capture these effects because it is linear by construction.
An alternative explanation is that the time series for the lending rate - which has been constructed by using a variety of sources - might present some ‘discontinuity' that affected the estimation results. This is a problem common to most euro area empirical studies: because long official time series are unavailable, they need to be replaced by synthetic measures derived from heterogeneous sources. Reassuringly, re-estimating the model using a shorter sample from just one source did not affect the estimation output to a great extent.
In this Appendix, we present three scenarios for 2010 and 2011:
Base case - loan growth comes back into positive territory at the end of this year, but remains barely above zero even in 2011. This scenario is based on our official GDP growth forecast for the euro area (1.2% in both years).
Optimistic case - loan growth picks up already in the middle of the year and continues to accelerate in 2011, but remains weaker than prior to the crisis. This scenario is based on a growth rate of 2.25%.
Pessimistic case - loan growth remains in negative territory until end-2011; credit to the private sector shrinks by about 2% in both years. This scenario is based on zero GDP growth in 2010 and 2011.
The simulated range of outcomes for loan growth suggests that a creditless recovery, i.e., a moderate expansion in GDP coupled with negative or barely positive loan growth, is quite possible. More specifically, our pessimistic scenario of flat GDP growth in 2010 is compatible with negative credit growth; our optimistic scenario of GDP expanding at the pre-crisis trend rate is compatible with a modest pick-up in credit growth.
Important Disclosure Information at the end of this Forum
Review and Preview
January 12, 2010
By Ted Wieseman | New York
The Treasury yield curve blew out to its steepest levels ever over the past week and inflation expectations in the TIPS market continued increasing to another round of new highs since mid-2008, as investors continued to worry about the Fed's seemingly excessive caution about the growth outlook and dovishness on monetary policy and inflation. Economic data through the week were somewhat mixed - good early indications for December consumer spending from the auto and chain store sales reports, upside in both ISM surveys, mixed results bearing directly on GDP that left our 4Q forecast at +5%, but a bit of disappointment in the employment report - and did little to change the market's view, with which we agree, that the economy has entered a sustainable though moderate recovery. The Fed's extremely easy monetary policy that was put in place to reduce the risks of a second Great Depression no longer seems appropriate to us for the greatly changed circumstances since mid-2009, and while few people think the Fed should be tightening aggressively certainly, we remain very concerned that there is no movement towards an exit strategy that begins to scale back the huge amounts of excess liquidity it has injected into the financial system. So the main impact of the employment report downside in the market - which largely seemed to be the result of bad weather, with continued improvement in claims confirming solid underlying improvement coming into 2010 - was frustration that it will probably add to the Fed's cautiousness and further delay policy movement, worsening inflation fears at the long end even as the short end was forced to continue reversing the big losses seen during the holidays. The short-end reversal was ongoing through the week. First, market reaction indicated that investors were surprised that the improved tone of the economic data, rising inflation expectations and roaring risk markets recently didn't prompt any hawkish shift in tone from Fed Chairman Bernanke or Vice Chairman Kohn in speeches Sunday. Then, the market was also apparently deeply surprised that the FOMC minutes actually reported more sentiment for further ramping up quantitative easing than for starting to reverse it. And then the employment report Friday prompted a completion of the near full reversal of the more hawkish Fed policy path priced into futures during the Christmas and New Year weeks. Market anxiety about the Fed, however, has been evident in about a 20bp sell-off in the long bond, lifting the curve to its historically unprecedented spreads, and 20bp rise in 10-year inflation expectations over the past three weeks even as Fed expectations have made this round trip and helped anchor short-end Treasury yields. We expect the ‘bond market vigilantes' to eventually help prompt the Fed to rethink its stance and move towards an exit strategy soon, with some language shift still possible at the upcoming FOMC meeting when policymakers will likely be contemplating the soon-to-be-reported 4Q GDP surge and what will probably be consensus expectations for a solidly positive January employment report.
On the week, 2s-30s spiked 24bp higher to 374bp and 2s-10s 15bp to 285bp, both all-time record high closes, as the front end surged on the dovish Fed, while the long end fell as inflation jitters remained elevated. The 2-year yield fell 19bp to 0.96%, 3-year 16bp to 1.52%, 5-year 12bp to 2.56%, 7-year 8bp to 3.30% and 10-year 4bp to 3.80%, while the long bond yield rose 5bp to 4.69%. Front-end upside came as dovish Fed rhetoric and the employment report forced a big Fed repricing in futures, with the Jan 11 fed funds contract surging 27bp to 0.935% after having sold off 32bp over the prior two weeks on rising expectations that the Fed might finally be ready to start moving to an exit strategy. Long-end weakness, on the other hand, came as inflation expectations continued rising. TIPS extended their run of big outperformance that has accelerated since the end of 3Q. The 5-year TIPS yield fell 18bp to 0.26%, 10-year 8bp to 1.35% and 20-year 3bp to 1.93%. This lifted the benchmark 10-year inflation breakeven another 5bp to 2.46%, another new high since mid-2008 after a 70bp increase since the end of 3Q (which has consisted of a 50bp rise in the nominal 10-year yield and a 20bp decline in the 10-year TIPS yield). TIPS continued to outperform into Friday even after commodity prices fell back somewhat after surging to new post-Lehman highs into Wednesday. Indeed, Seth Kleinman from our commodities strategy team notes that there seems to have been a notable shift developing recently in which energy prices are increasingly responding to rising inflation expectations in the TIPS market instead of the previously normal reverse pattern of causation. Mortgage market performance was mixed through the week, with the recovery that started before New Year after a terrible run through most of December extending into Wednesday before some ground was given up Thursday and Friday. For the week, current coupon MBS outperformed swaps and Treasuries slightly, with current coupon yields ending up close to 4.5%. The better recent tone allowed average 30-year conventional mortgage rates to fall slightly in this week to 5.09% after surging from a record low of 4.71% at the start of December to 5.14% in the last week of the month. Current coupon MBS yields at current levels should be consistent with 30-year rates holding near this 5.125% level. Note that while rates have risen off the lows, the plunge in home prices and the extended and expanded homebuyers' tax credit - worth about 50bp on rates for the carrying cost of an average house - have left housing affordability under conventional metrics at record highs. Swap spreads cranked down to record or near record lows Tuesday as a lot of well-received financial sector corporate debt issuance was being swapped. Some ground was given back over the rest of the week, but spreads still ended the week near historical tights, with the benchmark 10-year spread falling 2.5bp to 10.75bp compared with the all-time low of 8bp reached in May 2008.
Easy money for longer was taken as clear good news for now by risk markets, which ripped higher to start 2010 on an even stronger pace than the big gains seen in 2009. The S&P 500 gained 2.7% after rallying every day through the week, though the majority of the upside was posted Monday. The rally was led by energy, financials and materials, which all surged about 6%. Credit initially looked to be extending a recent significant outperformance versus stocks but stalled out midweek. Still gains were substantial for the week, with the investment grade CDX index 7bp tighter at 78bp, near the best level since late 2007, and the high yield index tightening 29bp through Thursday to 489bp and then rallying a bit further Friday. The leveraged loan LCDX index saw upside in line with high yield, with a 34bp tightening through midday Friday to 392bp. The commercial mortgage CMBX and subprime ABX markets surged higher early in the week but then faltered for a somewhat mixed showing for the week. The AAA and junior AAA CMXB indices only gained 1%, but the AA and A were up 3% and the AAA ABX index jumped 5%.
Non-farm payrolls fell 85,000 in December, and revisions to prior months were minor (though November was revised up a bit into slightly positive territory). The unemployment rate held steady at 10.0%, as the volatile household survey measures of employment and the labor force were both way down. Unusually cold weather during the December survey period appears to have significantly depressed job growth. The number of people saying they missed work because of bad weather, a useful proxy for weather impacts on job growth, was at the highest level for a December since 1977, and weather-sensitive areas like construction (-53,000) saw notably worse results. Our best guess is that payrolls probably would have been flat to slightly positive if the weather hadn't been so much colder than normal across most of the country during the December survey period. This should lead to a more positive weather impact in January, however, plus two leading indicators for job growth showed good results - temp employment (+47,000) posted another big gain for the largest rise on record over the past five months, and the average workweek held steady at 33.2 hours, a surprisingly solid result considering the weather, after a substantial increase last month. As a result of the workweek stability, total hours worked also held unchanged, a better-than-expected result. Combined with a 0.2% gain in average hourly earnings, the flat reading for total hours caused aggregate weekly payrolls, a proxy for aggregate wage and salary income, to rise 0.2%, extending a solid 3% annualized rebound in 2H09 and pointing to a modest rise in December personal income. A less negative weather impact, the continued sharp decline in jobless claims and very low numbers of layoff announcements in recent months in the Challenger survey that confirm solid underlying labor market improvement, some other technical factors (notably, potential seasonal swings in retail payrolls), and the beginning of census hiring point to a much better report next month. Sustainable job growth that we expect to average 130,000 a month over the next year is likely to begin in the January employment report, in our view.
Other early data releases for December were more obviously positive. Both ISM surveys showed upside, though the recovery continued to be led by the manufacturing sector. And results for motor vehicle and chain store sales were solid, pointing to another strong retail sales report. The composite manufacturing ISM index rose 2.2 points in December to 55.9, a high since April 2006. Most of the upside was accounted for by a big gain in the key orders index (65.5 versus 60.3) to a very strong level. Production (61.8 versus 59.9) and employment (52.0 versus 50.8) saw smaller increases. Also quite positive was a further decline in the customers' inventories index (35.0 versus 37.0) to another record low, with 37% of respondents saying they think their customers' inventories are too low versus only 7% who think they are too high. On the more negative side, expansion was less broadly based, with 9 of 18 industry groups reporting growth in December versus 12 in November. The composite nonmanufacturing ISM index rose to 50.1 in December from 48.7, moving just barely back into growth territory. Upside was led by the business activity index (53.7 versus 49.6). Employment (44.0 versus 41.6) was also less negative. The orders index (52.1 versus 55.1), on the other hand, pulled back, but remained above the 50 breakeven level. Growth by sector was less narrow than in November, with seven industries reporting growth in December, up from six, and nine contraction, down from 11. On consumer spending, motor vehicle sales rose to an 11.2 million unit annual rate in December from 10.9 in October, extending a big run-up in 4Q from the cash-for-clunkers payback in September. Meanwhile, chain store sales results were greatly improved, and while much of this reflected very easy base effects from the disastrous 2008 holiday shopping season, the numbers taken together also suggested solid sequential gains from November. With the modest upside in motor vehicle sales and an expected 0.4% gain in the key ‘retail control' component of retail sales, we see real consumer spending gaining a sold 0.3% in December. Mostly as a result of the weak starting point provided by the plunge in auto sales in September after cash for clunkers, consumer spending growth is likely to run at only +1.8% in 4Q. But a 3.7% annualized gain in the real PCE in the three months through December provides a much better ramp coming into 1Q.
There were a number of other releases directly impacting our 4Q GDP forecast released over the past week in addition to the consumer spending indications - construction spending, factory orders, state and local government employment, wholesale trade - but the results netted out and left our forecast little changed at +5.2%. The expected mix shifted a little more negatively, as we now see inventories adding a somewhat bigger +3.3pp and final sales gaining a slightly lower +1.9% (mostly as a result of weaker-than-expected construction spending results for November). Note that even with such a large expected inventory contribution, we still anticipate that inventories were liquidated at a substantial pace in 4Q, just a lot less so than the extreme rates seen mid-year. With demand showing modest growth, there is still significant catch-up to come in production relative to final sales over the next year.
The economic calendar in the coming week is on the light side and back-end loaded, with retail sales Thursday the most notable release. Supply will be the main Treasury market focus through most of the week, with four days of coupon auctions from Monday to Thursday - US$10 billion 10-year TIPS Monday, US$40 billion 3-year Tuesday, a US$21 billion reopening of the 10-year Wednesday and a US$13 billion reopening of the 30-year Thursday - and a bit more duration in the weekly bill auctions than normal with a US$26 billion 1-year auction Tuesday. Other data releases include trade Wednesday, the Treasury budget Wednesday, inventories Thursday and CPI and IP Friday.
* We look for the trade deficit to widen by US$1 billion in November to US$34 billion after the US$3 billion narrowing seen in October, with exports up 1.0% and imports 1.4%. Exports should be led by capital goods, in line with the decent gain in capital goods shipments figures, along with some price-supported upside in industrial materials. Upside in petroleum products, on both higher prices and volumes, is expected to account for the import gain, with port figures suggesting a pullback in non-energy goods imports after a big recent run-up.
* The budget deficit is expected to widen by about US$38 billion relative to the same period a year ago in December to US$90 billion. However, the bulk of this swing reflects timing differences that should be offset in January. The remaining deterioration in December amounted to about US$10 billion and reflected lower individual and corporate tax payments together with a special payment to Fannie Mae, higher unemployment benefits and increased interest expense. These factors were partially offset by a US$40 billion spurge in insurance premiums paid to the FDIC (part of the special acceleration in premiums that was announced back in the autumn). Assuming passage of the stimulus program that was approved by the House before the holidays, we see the F2010 budget deficit running at about US$1.25 trillion (or nearly 9% of GDP).
* We forecast a 0.5% gain in overall December retail sales and a 0.2% rise ex autos. Another solid gain at auto dealers is expected to be partially offset by a sharp weather-related drop at hardware stores and a price-related dip at gas stations. Meanwhile, the chain store reports pointed to mixed results for a couple of the discretionary categories, with general merchandise likely to register a gain while apparel outlets post another drop. The key retail control measure which feeds directly into the national income accounts is expected to be +0.4%. For the quarter as a whole, we see real consumption rising a modest 1.8%.
* A slight uptick in stockpiles at manufacturers, along with a sharper jump at the wholesale level, should help lead to a 0.5% rise in overall business inventories in November, which would be a second consecutive outright gain. Even after accounting for the anticipated rise in inventories, the I/S ratio should post another decline - from 1.30 to 1.28.
* We look for a 0.1% rise in the overall consumer price index in December and a flat reading ex food and energy. The food and energy categories are expected to register only slight upticks in December. Meanwhile, an unprecedented recent run of negative readings for owners' equivalent rent has offset a notable acceleration in prices for some other items and helped to keep the core CPI running right near +0.1% per month during 2009. This pattern is expected to continue in December. In fact, we look for the core to just barely round down to zero this month, with the year-on-year rate holding at +1.7%. Looking ahead, with housing market activity beginning to show signs of having reached a bottom, the shelter-related restraint on core inflation in 2010 is unlikely to be as significant as was the case in 2009.
* We forecast a 0.3% rise in December industrial production. The employment report pointed to a flattening out of factory output on the heels of a sharp jump seen in November. In fact, all of the expected rise in December IP is attributable to a weather-related spike in the utility component. The key manufacturing category is expected to be unchanged. By industry, we look for declines in motor vehicles, minerals, petroleum, furniture and printing to be offset by gains in metals, wood products, aerospace and chemicals. Finally, the utilization rate is expected to hit a new 12-month high.
Important Disclosure Information at the end of this Forum
Say Goodbye to the Job Recession...and Hello to the Census Effect
January 12, 2010
By Richard Berner, David Greenlaw, Ted Wieseman & David Cho | New York
Say Goodbye to the Job Recession (Richard Berner)
Job growth has yet to show up, but the great US employment recession is finally ending. The jury is still out on our thesis that employers went overboard in slashing payrolls and will start to hire back. At first glance, December's decline of 85,000 in non-farm payrolls seems to refute that notion, but we believe that weather-related headwinds played a significant role in the result. More important, several signs point to positive job growth very soon. We expect job gains over this year of just over 1% (130,000 monthly), along with a consistent rise in the workweek.
It's worth repeating that job and hours gains are critical for our call that growth will be sustainable through 2011. Rising jobs will provide the gains in income and confidence needed to support consumer spending. Rising income will also improve consumer creditworthiness and give lenders collateral comfort for further improvement in credit availability: It will reduce debt/income and debt service/income ratios, raise ‘cure' rates for delinquent mortgages, and help more consumers qualify for a loan. In short, rising employment will greatly reduce fears of a weak and faltering recovery.
While we don't think this recovery will be jobless, the pace of hiring is still likely to be moderate for three reasons: The economy itself still faces headwinds; companies remain determined to boost productivity; and uncertainty about labor costs, especially healthcare, may restrain hiring. Details follow.
Pent-up demand. Our thesis has been simple: Aggressive cuts to payrolls over the past two years have set the stage for a solid rebound, despite only moderate economic growth. What were minimal hiring excesses are long gone, and a growing economy has produced a hiring deficit. While there are several factors that will mute the hiring recovery, this pent-up demand will overwhelm them.
One measure of that pent-up demand is provided by a relationship between the economy and hours worked (and, with a projection for the average workweek, employment). The explanatory variables include the outlook for output, factors that affect productivity such as the services from capital, other variables aimed at capturing changes in trend productivity, and a dynamic adjustment process that captures the typical pro-cyclical surge in productivity early in recovery (see Appendix for equation). If positive, the cumulative differences between actual hours and those predicted by the relationship suggest that there is an overhang of labor to work off. As it turns out, the errors over the course of the expansion that ended in December 2007 were small, reflecting business caution about hiring. And through 2Q09, the errors cumulate to zero, suggesting that the aggressive job cuts seen in this recession eliminated any excess six months ago. We estimate that declines through 4Q have pushed those cumulative errors sharply negative, implying some underlying pent-up demand for labor that should materialize soon.
As we see it, the depth of the job recession and unsustainably strong productivity gains confirm this analysis. The economic recession and extraordinary cuts in jobs have taken the level of private payrolls about 610,000 below the trough of the previous recession in mid-2003, while the economy has since grown by about 11%. In addition, productivity has surged over the past nine months in time-honored cyclical fashion; such a surge is a hallmark of the first stages of recovery. But the pace is completely unsustainable. We estimate that non-farm productivity rose at an estimated 7% annual rate over the three quarters ended in 4Q09 - a pace last seen for brief periods in the 1960s. This hints that even moderate economic growth will trigger a pick-up in hiring.
Policy uncertainty may delay hiring. The main risk to our call for positive job gains is timing: Hiring always lags the recovery; the current surge in productivity growth speaks to that lag. This time, uncertainty around a variety of policy actions may further delay the pick-up. For example, we suspect that uncertainty tied to healthcare reform restrained hiring at small and medium-sized businesses for part of 2009. Both House and Senate plans would either mandate coverage or boost taxes. Why hire until the dust settles on those proposals? Employers are responding by boosting the workweek and hiring temps. The good news is that Democrats are determined to iron out differences between House and Senate healthcare bills, resolving this uncertainty. The bad news for employers is that, either way, hiring will cost more.
Four factors depressed hiring relative to the economy. Understanding why employers were so aggressive this time in cutting payrolls should help us to analyze the factors that will promote or inhibit recovery. Unfortunately, there is no simple answer. Yet it seems clear that the relationship between the economy and hiring has changed over time. For example, if Okun's Law (a rule of thumb relating economic growth and the unemployment rate) held from the past, given the peak-to-trough decline in the economy in 2008-09 of 3.8%, the unemployment rate should be about 8%, not 10%. Four factors probably changed the relationship: swings in employer-paid healthcare costs, increasing cyclicality of services industries, offshoring and demographics. In our view, these factors will continue to mute the hiring recovery, but will not preclude it.
1. Swings in healthcare costs. Swings in employer-paid healthcare benefits made hiring full-time workers more or less attractive over the past three decades. The tech bubble and strong economy fueled the late 1990s hiring boom, but controls that brought the growth in healthcare benefits down to zero may also have contributed to it. Since that time, Corporate America's hiring discipline, combined with a rapid escalation of healthcare costs, worked to correct those bubble-year hiring excesses, especially in manufacturing. Over the past 14 years, according to BLS data on employer costs for employee compensation, total compensation rose at a 3.4% annual rate. In contrast, health insurance costs rose at a 5.2% annual clip. To keep their benefits and take advantage of their being free of tax, workers have accepted lower growth in take-home pay.
2. More exposed - and thus more cyclical - services. Second, many traditionally stable services industries have become far more cyclical over the past two decades than in the past. Private services employment has ramped from 55% of total private payrolls to 83% in the past 50 years. In previous cycles, that shift reduced the cyclicality of the overall workforce. But the increased competition from deregulation and technological change has rendered many services industries (such as airlines, wholesale and retail trade, and IT) as cyclical as those in manufacturing and construction. And the financial crisis has promoted record declines in professional business and financial services, which accounted for 28% of private services payrolls at the end of 2007 and have shed 1.9 million jobs (7.2%) in the past 24 months.
3. Global competition. Third, a more open, globally exposed economy and the outsourcing and offshoring that goes with it have changed the dynamics of domestic hiring, although evidence for direct effects of offshoring on US employment is limited to perhaps 0.5%.
4. Coming demographic change. Fourth, the longer-term unemployment-economy relationship will probably change further as a slower-growing, aging population depresses labor force participation.
Metrics for improvement. Despite all these headwinds, several advance labor market indicators signal improvement in employment and hours. Notably, temporary help payrolls, often considered to be a leading indicator of labor demand, jumped by 166,000 in the past five months, the largest such rise on record. In addition, over the past few months the levels for overall payrolls were revised up significantly from what was originally reported; upward revisions are often a sign of improvement.
Other leading indicators are also looking up: Initial and continued claims for unemployment insurance benefits have declined steadily since peaking in June, although the decline in the latter overstates the improvement in labor markets as many unemployed workers have exhausted their regular benefits. The rise in federally funded emergency benefits has mostly offset the decline in regular jobless pay (see below Will the Real Jobless Tally Please Stand Up? for discussion). The employment components of the two ISM Indexes and stability in the private job openings rate over the past four months now seem consistent with gains in payrolls. Surveys of hiring and hiring plans such as those from our own Business Conditions Survey (the MSBCI) and Manpower, Inc. in the past three months improved noticeably (for more details, see Business Conditions: Encouraging Outlook for 2010, December 11, 2009). Importantly, factory regular and overtime hours have risen steadily since the spring. That is good news for income, which turned up 0.3% in November, and we expect hearty gains to continue in 1Q.
More stimulus coming. Improvement or not, today's 10% unemployment rate is likely to spur policymakers to consider new measures to stimulate job creation. For example, after the September jobs report, talk of enacting a new job tax credit surfaced in Washington. We believe that such a measure - if designed correctly - could be an important temporary source of effective stimulus. However, tax credits are politically unpopular. More likely is a package like the Jobs for Main Street Act of 2010, passed by the House in December. It included extension of COBRA insurance subsidies to 15 months, extension of unemployment insurance benefits, US$40 billion of transportation infrastructure outlays, US$5 billion for energy/water projects, US$27 billion for education grants, and US$2 billion for public housing. Such measures won't boost employment significantly, but they would be a plus.
Census a boon but still a moderate recovery. As Dave Greenlaw outlines below, hiring of census workers will boost payrolls temporarily in 2010, perhaps as soon as next month. Yet the underlying employment recovery is still likely to be moderate for three reasons: The economy itself still faces headwinds; companies are still determined to boost productivity in the face of the four factors discussed above; and uncertainty about labor costs, especially healthcare, may still restrain hiring.
Say Hello to the Census Effect (David Greenlaw)
To help conduct the census, the federal government hires hundreds of thousands of temporary employees. The census effect was not a factor in December's employment report, but it is likely to attract a good deal more attention over the course of the coming months. Every ten years, the US Census Bureau takes a snapshot of the population, determining how many people reside within the nation's borders, who they are, and where they live. The results of the census help to determine the make-up of the Congress and the apportionment of public expenditures.
Conducting the census requires the federal government to hire a large number of temporary employees. In fact, beginning with the 2000 Census, the number of workers was far larger than in the past (even after adjusting for the increase in the overall population). This reflected a court ruling which prohibited the Census Bureau from using statistical sampling techniques and adjustments that had been previously utilized to account for certain types of non-respondents. In other words, the Census Bureau now must perform a ‘hard' count - and can't rely on any estimates. The 2010 census forms will be distributed beginning in mid-March and the forms must be returned by April 21. The bulk of the census-related hires will show up in May when the door-to-door search for non-respondents gets underway. Most jobs will last for 6-10 weeks and pay US$10-20 per hour depending on the location.
We expect the number of census workers to build gradually in the months ahead and peak in May (note that the figures for 2010 are our own preliminary guesstimates). The BLS will provide a tally of the number of net new census workers in each monthly employment report, so the impact of this special factor can be quantified precisely - but only after the fact.
We derive our estimates using the Census Bureau's stated objective of hiring 1.2 million individuals to conduct the 2010 Census. We convert this to an estimated impact on payroll employment based on past experience. For example, the government hired 965,000 individuals to conduct the 2000 Census, and the maximum impact on the level of employment in any given month was 530,000. In the 1990 Census, the government hired 550,000 individuals, and the maximum impact in any single month was 335,000. The discrepancy between the announced number of hires and the impact on payrolls reflects the fact that not all of the positions overlap and there are probably some multiple job holders. So, for 2010, we converted the total targeted hiring of 1.2 million to a peak impact of about +700,000 on the level of employment. Then we phased in the hiring and separations according to our assessment of past patterns.
The unemployment rate is also likely to be affected by census workers, but it will be impossible to calculate the precise magnitude (even after the fact) because we won't know how many of the workers are new entrants to the labor force and how many are multiple job holders. However, we can provide some rough parameters. Under the extreme assumption that all census workers are job finders (i.e., individuals previously counted as unemployed), the impact on the unemployment rate would peak at about -0.50 percentage point. Under the more realistic assumption that about one-third of the workers are job finders, one-third are new entrants and one-third are multiple job holders, the peak impact on the unemployment rate should be only about -0.15 percentage points. Of course, these effects will eventually wash away, and if past patterns hold, we should get relatively ‘clean' readings for the unemployment rate by October.
Will the Real Jobless Tally Please Stand Up?
The number of people receiving benefits under regular unemployment insurance programs has dropped sharply from its peak of 6.9 million at the end of last June, signaling improvement in labor markets. This standard continuing claims series declined to a seasonally adjusted level of 4.802 million in the week ended December 26 - the lowest level since January last year. However, this metric clearly overstates the improvement, because many workers have exhausted their regular unemployment benefits.
Additional series track extended benefits offered to such workers under two federal programs. The number of people receiving benefits under such programs has surged to 5.4 million in the week ended December 19 (these data are not seasonally adjusted and are reported with a lag of an additional week) from zero last January. In the week ended December 19, the number of benefit recipients under the two extended programs rose by 165,000, and the net increase was revised sharply higher in the preceding week - from 199,000 to 658,000.
It is clearly important to account for both of these developments to assess the state of labor markets. Indeed, the opposing movement of continuing claims compared with Emergency Unemployment Compensation (EUC) benefits makes it tempting to add the two together to arrive at a total pool of unemployed. The result would be a much more dire picture of job market stress and slack in labor markets. But there are two reasons why such a procedure is invalid, and such a total would significantly overstate the current weakness in labor markets:
1. The extended benefits series aren't seasonally adjusted, and there is typically a seasonal surge in joblessness around the end of the year. While it is not possible to adjust these data accurately for seasonal variation because the trend is unknowable, the unadjusted data may seriously overstate the weakness. Using the seasonal factors for the standard series is also inappropriate; the seasonality in the two benefits series is not the same because the trend component of the two series is completely different. (The calculation of seasonal adjustment factors involves the decomposition of a series into its trend, seasonal and irregular components, and if the trends are different, the multiplicative seasonal factors will be different.)
2. More important, the EUC data overstate the weakness in labor markets because they are affected by ongoing increases in the duration of benefits. To start, the EUC program offers benefits to people who were unemployed and not covered by the standard UI program. For example, Congress retroactively extended eligibility to people who were out of work in 2006. Moreover, in most states, recipients are now eligible for 99 weeks of unemployment benefits. Even in the best of times, about 33% of recipients exhaust all the benefits that are available to them. This suggests that there is a sizeable ‘underground' economy in which beneficiaries will extract all available benefits. So, the extended series is unlikely to start to come down until well after the labor market has turned the corner.
The bottom line: Neither the standard continued claims data nor the extended series provide any real information content at this point and should be ignored. All of the focus should be on initial claims.
Appendix
The equation we used to derive the estimated labor overhang predicts the change in hours worked in private non-farm business. Errors from the equation are cumulated and scaled by hours to obtain the overhang estimates.
The equation closely follows that of Macroeconomic Advisers; see their study Productivity and Potential GDP in the ‘New' US Economy. They begin with the assumption that output is a function of capital and labor inputs in a constant-returns-to-scale production function. Labor productivity - the ratio of output to labor input, or hours - can be decomposed into three components: technological improvement, which are assumed to be exogenous, the rate of growth of capital services relative to the growth of labor inputs scaled by capital's share, and other cyclical factors. Normalizing the equation on the growth of hours worked makes them a function of the growth of output, capital services and other factors.
Important Disclosure Information at the end of this Forum
Disclosure Statement
The information and opinions in Morgan Stanley Research were prepared by Morgan Stanley & Co. Incorporated, and/or Morgan Stanley C.T.V.M. S.A. and their affiliates (collectively, "Morgan Stanley").
Global Research Conflict Management Policy
Morgan Stanley Research observes our conflict management policy, available at www.morganstanley.com/institutional/research/conflictpolicies.
Important Disclosure for Morgan Stanley Smith Barney LLC Customers
The subject matter in this Morgan Stanley report may also be covered in a similar report from Citigroup Global Markets Inc. Ask your Financial Advisor or use Research Center to view any reports in addition to this report.
Important Disclosures
Morgan Stanley Research does not provide individually tailored investment advice. It has been prepared without regard to the circumstances and objectives of those who receive it. Morgan Stanley recommends that investors independently evaluate particular investments and strategies, and encourages them to seek a financial adviser's advice. The appropriateness of an investment or strategy will depend on an investor's circumstances and objectives. Morgan Stanley Research is not an offer to buy or sell any security or to participate in any trading strategy. The value of and income from your investments may vary because of changes in interest rates or foreign exchange rates, securities prices or market indexes, operational or financial conditions of companies or other factors. Past performance is not necessarily a guide to future performance. Estimates of future performance are based on assumptions that may not be realized.
With the exception of information regarding Morgan Stanley, research prepared by Morgan Stanley Research personnel is based on public information. Morgan Stanley makes every effort to use reliable, comprehensive information, but we do not represent that it is accurate or complete. We have no obligation to tell you when opinions or information in Morgan Stanley Research change apart from when we intend to discontinue research coverage of a company. Facts and views in Morgan Stanley Research have not been reviewed by, and may not reflect information known to, professionals in other Morgan Stanley business areas, including investment banking personnel.
To our readers in Taiwan: Morgan Stanley Research is distributed by Morgan Stanley Taiwan Limited; it may not be distributed to or quoted or used by the public media without the express written consent of Morgan Stanley. To our readers in Hong Kong: Information is distributed in Hong Kong by and on behalf of, and is attributable to, Morgan Stanley Asia Limited as part of its regulated activities in Hong Kong; if you have any queries concerning it, contact our Hong Kong sales representatives.
Morgan Stanley Research is disseminated in Japan by Morgan Stanley Japan Securities Co., Ltd.; in Canada by Morgan Stanley Canada Limited, which has approved of and takes responsibility for its contents in Canada; in Germany by Morgan Stanley Bank AG, Frankfurt am Main, regulated by Bundesanstalt fuer Finanzdienstleistungsaufsicht (BaFin);in Spain by Morgan Stanley, S.V., S.A., a Morgan Stanley group company, supervised by the Spanish Securities Markets Commission(CNMV), which states that it is written and distributed in accordance with rules of conduct for financial research under Spanish regulations; in the US by Morgan Stanley & Co. Incorporated, which accepts responsibility for its contents. Morgan Stanley & Co. International plc, authorized and regulated by Financial Services Authority, disseminates in the UK research it has prepared, and approves solely for purposes of section 21 of the Financial Services and Markets Act 2000, research prepared by any affiliates. Private UK investors should obtain the advice of their Morgan Stanley & Co. International plc representative about the investments concerned. RMB Morgan Stanley (Proprietary) Limited is a member of the JSE Limited and regulated by the Financial Services Board in South Africa. RMB Morgan Stanley (Proprietary) Limited is a joint venture owned equally by Morgan Stanley International Holdings Inc. and RMB Investment Advisory (Proprietary) Limited, which is wholly owned by FirstRand Limited.
Trademarks and service marks in Morgan Stanley Research are their owners' property. Third-party data providers make no warranties or representations of the accuracy, completeness, or timeliness of their data and shall not have liability for any damages relating to such data. The Global Industry Classification Standard (GICS) was developed by and is the exclusive property of MSCI and S&P. Morgan Stanley bases projections, opinions, forecasts and trading strategies regarding the MSCI Country Index Series solely on public information. MSCI has not reviewed, approved or endorsed these projections, opinions, forecasts and trading strategies. Morgan Stanley has no influence on or control over MSCI's index compilation decisions. Morgan Stanley Research or portions of it may not be reprinted, sold or redistributed without the written consent of Morgan Stanley. Morgan Stanley research is disseminated and available primarily electronically, and, in some cases, in printed form. Additional information on recommended securities/instruments is available on request.
The information in Morgan Stanley Research is being communicated by Morgan Stanley & Co. International plc (DIFC Branch), regulated by the Dubai Financial Services Authority (the DFSA), and is directed at wholesale customers only, as defined by the DFSA. This research will only be made available to a wholesale customer who we are satisfied meets the regulatory criteria to be a client.
The information in Morgan Stanley Research is being communicated by Morgan Stanley & Co. International plc (QFC Branch), regulated by the Qatar Financial Centre Regulatory Authority (the QFCRA), and is directed at business customers and market counterparties only and is not intended for Retail Customers as defined by the QFCRA.
As required by the Capital Markets Board of Turkey, investment information, comments and recommendations stated here, are not within the scope of investment advisory activity. Investment advisory service is provided in accordance with a contract of engagement on investment advisory concluded between brokerage houses, portfolio management companies, non-deposit banks and clients. Comments and recommendations stated here rely on the individual opinions of the ones providing these comments and recommendations. These opinions may not fit to your financial status, risk and return preferences. For this reason, to make an investment decision by relying solely to this information stated here may not bring about outcomes that fit your expectations.
|
![](/img/clear.gif"){kind=spacer}