U.S. July Payrolls Indicate the Worst of Labor Market is Yet to Come

Today’s employment report showing large job losses and a spike in the unemployment rate indicate things are only getting worse for the U.S. consumer, especially when he is dealing with high cost of living, mortgage and credit card debt and tighter credit conditions. Given the jobless recovery and slow wage growth since the last recession, households used their homes and credit borrowing to finance consumption and drive the economy. But the housing and financial sectors took a hit and now with even the labor market flashing a ‘no-entry/only exit signal’ at them, it seems like a perfect storm for the U.S. households.

What are the Indicators Showing?

Several leading labor market indicators are now showing trends that in the past have been observed only when the economy was in the initial stages of a recession. After total job losses of 412,000 in H1 2008, July witnessed job loss of 51,000 in July. Payrolls often called lagging indicators ‘usually’ start softening as the economy slows and turn negative when the economy enters a recession, moving back to the positive territory only after the recession ends and recovering slowly thereafter. Job creation started slowing since mid-2007 but has turned negative since Jan 2008. Job growth has in fact turned negative on a 3-month moving-average and quarterly basis since Feb 2008. Job growth on a 12-month moving-average and yearly basis has also slowed starting Jan 2008.

Data Source: Bureau of Labor Statistics

The unemployment rate started inching up since Dec 2007 and hit a four-year high of 5.7% in July. The unemployment rate usually tends to rise steadily as the recession deepens, peaking nearly two years later and recovering very slowly, often reaching the pre-recession level only after four years. The number of unemployed as a share of total population and the unemployment rate among laid-off workers claiming for jobless benefits has also shown an upward trend in recent months. The latter called the jobless rate rose to a high of 2.5%. The increase in unemployment rate is especially large (like during most recessions) for teens, blacks and Hispanics. Looking at the state level trend, it is not surprising that states hit hard by the housing crisis have shown a significant increase in the unemployment rate (as of June-08): California (6.9%), Nevada (6.4%) and Florida (5.5%); so have states affected by auto sector woes: Ohio (6.6%), Michigan (8.5%). However, booming exports and energy prices may have helped keep the unemployment rate down in states like Texas (4.4%), Louisiana (3.8%) and Oklahoma (3.9%).

However, due to underlying trends in the labor market, the unemployment rate may be hiding or understating the extent of labor market deterioration. This is because usually, the hiring rate turns flat as the economy enters a recession and continues to remain subdued or declines for almost two years before recovering (Several studies have in fact highlighted the importance of slowdown in hiring relative to the pick-up in lay-offs during a recession). This time too, it is the significant slowdown in hiring (of both unemployed workers and new labor force entrants) that is contributing to the labor market weakness. Hiring started slowing since Q3 2007 and declined to 3.2% in May as firms facing uncertain economic prospects and rising production costs off late have been less eager to hire workers. This has deeply affected job prospects for fresh graduates and summer-job seekers, including school/college students and teens. Similarly, job opening has slowed down since Q4 2007 and was at 2.6% in May. So owing to sharp slowdown in hiring, several new and laid-off workers unable to find jobs are backing out of the labor force (called discouraged and marginally attached workers) or working part-time (as firms prefer them over full-time workers during economic slowdown). In fact, the labor participation rate has been flat or somewhat weak since early-200, thus keeping a lid on the unemployment rate. Also given the very modest job creation since the last recession, firms may actually have less excess labor to shed, preventing employment from declining severely and thereby holding down the unemployment rate. So instead, calculating the unemployment rate including these workers (long-term unemployment rate) shows a continuous upward trend since Dec 2007, hitting a double-digit figure of 10.3% in July.

One of the most important indicators of the labor market strength are the initial and continued unemployment insurance claims (UI) since they usually start rising even before the economy enters a recession. While both started spiking since Mar-Apr 2008 and the latter hit recessionary levels of over three million in mid-April while the former started jumping into the +400k level only since June-end, which implies that slower hiring/rehiring of workers were initially driving the labor market weakness. As lay-offs are gaining pace, initial claims are expected to rise further in the coming weeks. Moreover, continued claims may be underestimated as workers who are unemployed for over 26 weeks (amid weak hiring prospects) and exhaust their benefits are not counted in the data Those eligible for extended jobless claims under the Congress Bill are accounted for in a different dataset. Also, self-employed workers whose share in affected sectors like real estate, rental and leasing and construction has risen in recent years are also not eligible for jobless claims. Moreover, many of the lay-offs are occurring in the high-paying financial sector and the illegal immigrants-laden construction sector where workers may not file for claims. Others may also delay or decide not to claim benefits in the hope of finding a new job soon even as different states have varied eligibility requirements to file for claims.

Apart from jobless claims, other ‘leading indicators’ have been weakening since H2 2007. The Index for newspaper and online job ads for Help-Wanted has been declining since Q4 2007 with the former hitting a record low in July and the latter down 7.9% y/y. The Monster Website online job availability Index is also down 14% y/y. The Manpower Index also indicates that firms planning to increase hiring in Q3 2008 is at a five-year low. The Conference Board Employment Index in June continued to reflect the fact that households are finding jobs hard to get and see fewer jobs available in the next six months. The weak demand for temporary workers escalated after Jan 2008, declining further by 29,000 in July. Also, weekly hours of labor started flattening since Jan 2008 and hit a record low and a recession level in July. In addition, part-time workers have increased since Q3 2007 and while full-time workers started to decline since Dec 2007.

Apart from the fact that the BLS data is subject to large revisions, there has been much criticism of the BLS methodology and its inaccuracy in predicting business cycles due to which employment numbers would indicate a recession only after it has begun. For firms not covered under the survey, BLS estimates how many firms (jobs) come into business and go out of business each month (called the birth/death model). But this model often overestimates the former and underestimates the latter during an economic slowdown, thus missing the turning point in business cycles. In fact, in the recent months, BLS has been overestimating job creation in the weakening sectors like construction, leisure and hospitality, finance and thus underestimating the extent of job losses.

Sector-Wise Job Losses:

Private sector employment started slowing in 2007 and turned negative in Dec 2007 losing 665,000 jobs including a loss of 76,000 jobs in July. As expected, job losses first showed up in real estate, construction, manufacturing, financial services and related activities. The manufacturing sector being highly cyclical, starts shedding jobs when the economy enters a recession lost 35,000 jobs in July (271,000 ytd) led by goods-producing sector (46,000 in July and 522,000 ytd). While the ISM manufacturing employment Index began showing weakness since end of 2007, the ISM Non-Manufacturing Index also started softening since Jan 2008 and both contracted in June.

Automobile, airlines, shipping sector woes (on slowing consumer demand) have been exacerbated by rising fuel costs; technology sector also expects lower demand from consumers as well as firms (as firms cut IT spending as one of the first steps to cut costs). While many firms have been bullish about the booming global economy and therefore manufacturing exports offsetting lower demand in the U.S, it may just be a matter of months before the U.S. slowdown and subdued demand spreads to the rest of the world and hits export-oriented manufacturing industries. Correction in oil and commodity prices will also impact employment in these sectors. Self-employment which boomed during 2004-07 has also shown a decline since July 2007 but by being counted only in the household survey, the payroll numbers ignore job loss among this group

Construction employed after boom during 2004-06, particularly in the residential and specialty trade contractor sectors) has witnessed large job losses in the recent months even without taking illegal immigrants into account. In July it lost 22,000 jobs taking the Jan-Jul figure to 290,000. Residential construction employment after peaking in 2006 started deteriorating in H2 2007 (especially in residential specialty trade) and shed 14,000 jobs in July. It was expected that the non-residential sector (where job creation was going strong in 2006 and early-2007 even when residential employment had started slowing) would provide some cushion by absorbing labor from the residential sector but here too employment has been falling since Q3-07 albeit at a slower pace. Given the pending correction in home prices and inventories, construction and housing-related employment will continue to trend down in the near future. Recent trends in commercial real estate show that jobs there too might be at risk. There has been some evidence of slowing remittances to Mexico as immigrants lose jobs and also slowing inflow of new immigrants into the country (and even outflow of some immigrants) on weakening job prospects. However, booming agriculture prices may help absorb some immigrants laid-off in the construction sector into the farm sector.

Services have been a bright-spot in terms of job creation in the recent decades. They tend to shed jobs later in the recession (due to the lag effect in consumers feeling the pinch and cutting down on spending). As a result, job creation in services has been weakening since Jan 2008 and we saw job loss of 5,000 in July. This may be a matter of concern given its high share in GDP and employment. Retail and business and professional service employment have also shown an upward trend over the years and have tended to lose more jobs in the last two recessions compared to the previous ones (due to increasing significance of consumers in the economy). Retail employment started slowing in mid-2007 and showed continued job loss of -17,000 in July, while employment in business and professional services (which started declining from Jan 2008) showed a decline of 24,000 in July (-189,000 ytd).

Government, and health care & education jobs have been resilient since the last two recessions showing little weakness. In fact, these sectors have been significant sources of job creation in the recent years (especially in govt sector compared to the private sector) and continued to add 25,000 and 39,000 jobs in July respectively. However, it will be interesting to see how long government jobs can hold up especially the state and local government ones given that several state governments and municipalities are facing huge deficits and are planning to cut down on services like health, education and infrastructure spending (even as falling individual and corporate income, property prices and consumer spending is eroding their tax revenues, and higher interest cost on muni bonds is raising their debt for health care, transportation and schools).

Financial sector jobs were unchanged in July from June. Given that this sector has taken a major hit from the credit meltdown one would have expected much larger lay-offs (especially at Wall Street) than what the numbers are showing. But this may be because only a small sub-sector of the financial sector has been affected by the credit crisis (like credit intermediation, fixed income, investment banking, asset management, real estate brokerage and mortgage division) and is shedding jobs while other sub-sectors like equities, currencies and commodities have been holding up (so far). Moreover, the affected sectors represent a small share of the total financial sector jobs and total Wall Street jobs. Wall Street represents a small share of the total financial sector jobs and many workers offered severance packages continue to be counted in the payroll survey. Credit intermediation (-3,800 in July), finance and insurance, real estate (-2,400 in July), rental and leasing services have also shown large decline in employment since early-2007 (especially during Aug-Oct 2007). Commercial banking also witnessed weak to negative job growth during this period. Over 100,000 finance jobs have been lost since Aug 2007 with over 21,000 of them since Jan 2008. While Wall Street has lost only over 10,000 jobs so far, close to 35,000 lay-offs are expected in the coming quarters but these estimates have been raised up recently following continued bank writedowns and job cut announcements. As the credit crisis spreads to the other sub-sectors of the financial sector (as liquidity crunch reduces capital flowing into asset management, hedge funds, private equity, consumer/auto loans, and banks and fund managers suffer credit losses and begin cutting operating costs), job losses are also bound to spread. Moreover, as the global economy slows and oil and commodity boom end, the presently booming sub-sectors will also feel the pain. In fact, job creation in securities, commodities, investment firms started slowing since late-2007 though is still positive.

Not many sectors have added excess labor since the last recession (or at least at the same pace as in previous recoveries) to now shed workers. This is particularly true for the manufacturing sector (and partly service sector) where several jobs have moved offshore since 2001. However, some sub-sectors especially in construction and finance sectors have seen strong hiring in recent years, making it obvious for them to shed jobs once the bubble bursts.

Implications:

So far, firms had perceived the crisis to be sector-specific and a short-lived one therefore making it inefficient and costly to fire and later re-hire workers and were instead hiring fewer workers, reducing overt-time, full-time and temporary workers and relying more on part-time workers. They have also been making the existing workers work for fewer hours or adjusting their wages, perks and bonuses. However, now the firms realize that weakness in consumer spending is going to get worse in H2 2008 as consumers face increasing financial headwinds and the impact of tax rebates fade. And add to this the spike in fuel and raw material prices and its impact on production costs. Indicators for industrial production, manufacturing activity, capex, factory and durable goods orders also show that producers plan to scale back on production As a result, of slowing economy and rising costs, firms have lowered their earnings forecast and plan to cut operating costs and lay-off thousands of workers in the coming months. Hence, lay-offs are bound to escalate ahead and add to the pool of unemployed workers so far dominated by slow hiring trends. Also after spending the rebates, consumers will continue to face job loss, slowing asset and wage income, high debt, credit constraints, and inflation (food, fuel prices and prices of other goods as well since most firms plan to pass on higher costs to consumers). So consumer spending on goods and services will only deteriorate ahead, so that job losses will spill-over to other sectors (with a lag) creating a vicious circle. Impact on consumption may especially be true when job cuts occur in the high-paying service sector and some high-skilled sub-sectors like financial and business and professional services. Also during a recession, workers (particularly in less-skilled and manufacturing sectors) find new jobs where wages are relatively less compared to their previous job. Job losses in retail (consumer durables, luxury goods, home-related goods, apparel, textile, electrical sectors), leisure and hospitality, business and professional services are bound to intensify as consumers opt for discounted and low-ticketed items. The diffusion index also shows that job losses started spreading to other sectors of the economy starting Nov 2007 and this trend continued even in July as economic weakness slowly spread from residential to commercial real estate, financial to non-financial corporate sector. A consumer-centered recession may exacerbate this and pose further downside risks to a prolonged economic slowdown and recovery.

However, there doesn’t seem to be any significant risk of wage pressures that might toughen Fed’s job of price stability by creating a wage-price spiral. Average hourly and weekly wages have stabilized in recent months (the former rose only 0.3% in July while the latter was unchanged). Unit labor costs and compensation have also weakened since Q4 2007. The employment cost index in Q2 2008 rose slowest in two years at 0.6% while slowing wages and benefits led to a weaker compensation growth of 3.1% y/y. But more importantly, productivity growth has rebounded since H2 2007 (which may be because firms started using fewer labor or labor hours as the economy began slowing). Anyways, compensation growth (which in recent years has mainly been driven by bonuses, perks and stock options) is bound to decline as firms cut costs. Moreover, the Fed’s past worries of shortage of skilled labor in certain states and sectors (say finance) is fading as demand and even compensation of skilled labor is softening while productivity growth continues to be buoyant. Also, as producers increasingly pass on higher prices to consumers, real wages may remain subdued and prevent the risk of a wage-inflation spiral. However, this also implies that labor is taking home less in real wages. In fact, real hourly compensation was negative in Q1 2008 and real wages have been negative since Oct 2007. Labor and individual income tax receipts as well as withholding taxes have also shown weakness recently. So while somewhat good from the monetary policy perspective, but as long as inflation erodes the purchasing power of workers, it may tend to delay the recovery in consumer spending and hence economic growth. And firms recognizing weak consumer demand may reduce production activity and workforce, thus re-enforcing the labor market weakness.

It is argued that given the structural changes in the economy and the ‘great moderation’ and also the demographic and compositional changes in the labor market in recent decades (including rising employment share from manufacturing to services, declining labor participation rate and labor force), the extent to which the unemployment rate and job losses rise during a recession may be contained. Also, like the last recovery, many manufacturing and service jobs may again move overseas (as a part of re-structuring and cost-cutting by firms). Labor participation for teens and women may decline further if labor market weakness prolongs. Older workers depending on their skill level may or may not return to the labor force. But a prolonged recession and/or a slow recovery might lead to significant job losses, pushing up the unemployment rate to higher than forecast (which itself was revised up by the Fed in June) up to 6.5-6.7% levels by 2009. Taking cues from past recessions, the unemployment rate may start declining by 2010 but may recover to pre-recession level only by 2012.

For a more detailed analysis of the U.S. labor market take a look at the RGE Writing: U.S. Labor Market Dynamics in a Slowing Economy (Restricted Access)