| |
 May 2004 Job Hunt
Ever since then other economic measures have also signaled renewed expansion. Productivity and corporate earnings have shown marked increases while consumer confidence has remained at high levels. Capital spending appears to be reviving while layoffs are diminishing. As you'd expect, the stock market has benefited from all of this. Yet at least until March, job growth was surprisingly anemic. This isn't to say the unemployment rate has been high; it hasn't. It's hovered just under 6% for quite some time now. In fact it never really spiked as it has in previous recessions.
Some argue that the reported unemployment rate actually understates the numbers since so many displaced workers have given up their job search. This could very well be the case in light of the fact that unemployment duration -- the number of weeks it takes displaced workers to find a new job -- is at near-term highs. Regardless of the accuracy of specific reporting, no one disputes that this recovery has lacked job growth. If you haven't heard this yet, you most certainly will since politicians are seizing on this as an election year issue. But putting political rhetoric aside, why have employers been so slow to hire new employees? Is this some sort of anomaly in the current recovery or can it be explained in the context of those that have gone before? The Usual SuspectsAll economic indicators don't move in tandem. If they did, there would be a lot more certainty in the economic "science".Instead, some tend to be leading indicators (e.g. the stock market) that move before the economy, lagging indicators (e.g. employment growth) that move after the economy, while others are "coincident" indicators that move along with the economy. Generally, leading indicators can be used to predict the direction of the overall economy while the others confirm (or don't confirm) the prediction. There's always a delay between economic developments and when the results are reported. As a result, the economy often turns well before even leading indicators send their signals.
Things often look their worst at the beginning of a recovery. At that point the economy has been slowing for months. Consumers realize this and lower their expectations, perhaps even fearing for their jobs. As you'd expect, consumer confidence is low as recovery begins. Consumers are naturally much more reluctant to spend, particularly for those goods and services that are considered non-essential. As a result, businesses are stuck with excess inventories and have little need for additional production. With no demand for existing inventory, there's no need to spend on new plant or equipment, much less new hires. Businesses may actually cut back on employment in an effort to save costs and boost the bottom line. This option helps feed the vicious cycle by further reducing consumer confidence. But at some point, this cycle is broken and that's the turning point from recession to expansion. New products or technologies may spur demand. Perhaps international developments or geopolitical events spark economic activity through new military or humanitarian spending. In many instances, it might just be the result of the excesses of the previous boom being worked out of the economy. Whatever the catalyst, production and demand eventually pick up. Businesses may have run through existing inventory and now need to resume production. This may require the purchase of new equipment or addition of staff. At the very least, it will reduce layoffs. Consumers confidence grows with firmer job prospects and increasing incomes. Demand for goods and services -- even nonessential ones -- grows and the vicious cycle becomes a virtuous one. It may take weeks, even months, before reported economic figures reflect renewed activity, especially at the turn from recession to recovery. By the time it's confirmed by the numbers, the recovery is usually well under way. Different This Time?So what's different this time? How can everything else point to recovery except job growth? Is it no longer an essential piece of the puzzle? To answer these questions, let's look back at other economic series have historically told us about employment growth.
The Gross Domestic Product (GDP) certainly goes along with the consensus view that the recovery turned in early 2002. Until that point it had actually been shrinking, yet it reversed course for good in early 2002. So how has GDP growth been related to employment growth? Looking back over the past 22 years (the extent of Baseline's data), the two are, as you'd expect, directly related. It's intuitively reasonable to expect employers to be hiring as their output increases. But when does the hiring occur? Does it begin as soon as there's a ramp up in sales, does it come before, or perhaps after? Using Baseline's data and regression analysis, while the two are always closely tied, the strongest relation occurs when GDP growth leads by 7 months (or conversely, when employment growth lags by 7 months). This is illustrated in Chart 1. Not only is this visually compelling, it's also statistically significant with the correlation coefficient of .72. (For an brief explanation of these concepts, see The Quantitative Approach.) This actually makes intuitive sense, too. In essence it says that the GDP begins rising well before firms begin hiring. Put yourself in the position of a business owner: Throughout the recession, you've seen little demand for your widgets. In fact, you've probably suffered through an inventory backlog. Now your distributors are suddenly ordering more widgets -- thank goodness! So what do you do? Do you immediately bring in new employees to ramp up production? Probably not. It would make much more sense to run through your inventory first. That will give you the opportunity to determine if the increased demand will persist before committing more resources to production. More importantly, selling widgets out of inventory will be better for your bottom line. You've already incurred the cost of their production so now their sales will go right to profits. There's no need to hire new staff for production when you already have inventory on hand. Even when you've depleted your inventory you still might may have no need to hire. If your present workforce can produce enough widgets to meet demand, there's no reason to incur the costs of more employees. It's only when demand exceeds your ability to produce that you'll add staff. The data of Chart 1 suggest this has commonly occurred 7 months after a resumption in GDP growth. Don't get too hung up on the seven months aspect, that's just a statistical average. The important point is that hiring has followed GDP, not preceded or even run with it. In the average recovery, you could expect employment growth to pick up well after the turn in GDP. Of course that turn most recently occurred in early 2002, yet there have been few signs of employment growth until almost 24 months later, in March of 2004. Is something else at work here?
Indeed there is. To see why, go back to the widget example. Again suppose demand has picked up, but now you've run through your backlog. If you have 10 employees and each is producing 10 widgets a day, you still won't have to hire anyone until demand exceeds 100 widgets per day. OK, suppose demand does go to 110 widgets per day. Do you hire a new employee? Not necessarily. Why? First off, if each existing worker isn't working a full capacity -- if each can make at least one more widget per day -- you can meet existing demand without incurring the cost of a new hire. So this is initially a question of capacity. Like your widget factory, all employers will want to make sure they are producing to their full capacity before considering increasing it. Secondly, even if you're at full capacity, it's still not necessarily time to hire. It may be time to add resources, but not necessarily human ones. Again at the widget factory, if 10 widgets per day is all each worker can be expected to produce, you'll have to increase capacity to meet the demand for 110 per day. Ignoring the costs of taxes and benefits (a big assumption), if you pay your employees $10 an hour to work 40 hours a week, it will cost you $20,800 per year to add one new employee to make those 10 extra widgets per day. With this approach, you've incurred an additional $20,800 expense to make the additional widgets. But if you could spend less than $20,800 to buy updated equipment or technology that would enable each of your present employees to make just one more widget a day, it would increase your bottom line more than by hiring a new worker. In fact, if such a purchase had a useful life of more than one year, you still might consider it even if it cost more than $20,800. Increased productivity and efficiency enables firms to produce more goods with fewer resources and associated costs. Unfortunately, some of those cost savings come at the expense of jobs. There's no need to create new jobs if increased efficiency allows sufficient production at a lower cost. Actual ResultsChart 3 shows the relation between employment growth and manufacturing productivity. It suggests there's an inverse relationship: When one rises, the other falls. In other words, when productivity is on the increase, employment growth is declining and vice-versa.If you stop to think about it, it makes sense -- at least over the short-term. As long as you can squeeze more productivity out of your current employees, you won't be adding more. Fully utilizing your present resources is simply more efficient. But when you can no longer do this, you must take the more "inefficient" route and add to your payroll. It's only when productivity gains begin to top out that hiring becomes a viable alternative.
That's exactly what Chart 3 indicates. The best fit of the two series is when employment growth leads by two months. In other words, employers start to hire just as productivity begins to level off. This is precisely what we've been seeing in the present recovery. Employers spent the latter part of the '90s investing in new equipment and technologies not only to meet demand but to prepare for Y2k. As soon as the millennium turned, the Internet bubble burst and the economy turned south leaving employers with all that new capacity but greatly diminished demand for their products. As the economy finally turned in early 2002, demand returned. Manufacturers were able to increase output by taking advantage of their unused capacity (see Chart 4). Productivity increased dramatically (look back at Chart 3) but at this point, there was no need to add new jobs. In the past few months, productivity, while still growing, has climbed at a slower pace. If the January-March jobs reports are to be believed, this slowdown is coming just as employment is beginning to pick up. This is exactly what you would expect. Indeed, one could plausibly argue that job growth has been late arriving in this recovery precisely because productivity growth was so strong. The increased efficiencies from new and improving technologies while a boon to employers' bottom lines has worked to delay the need for new workers. So maybe this recovery really isn't so different from those that went before. In fact, it might just be offering a glimpse of what future recoveries will look like. As the U.S. economy becomes more service rather than manufacturing oriented, technological improvements will lead to greater efficiency while lengthening the time between the start of recoveries and the need to add more workers. There aren't too many economists out there on the campaign trail so don't expect such facts to get in the way of any good sound bites. Unfortunately pointing fingers and placing blame is still a good way to get elected. But now you at least know better. Search this site! Just enter you key word or words:
Get current quotes or follow your own custom portfolio,
courtesy of E-Line Financials:
|
||||||||||||||||||||||||||||
ITH 20-20 HINDSIGHT, MANY ECONOMISTS contend the most recent recession ended in (or at least around) January 2002. Some argue it was actually ending in late summer, but the events of 9/11 set it back several months. Regardless of the exact date, GDP growth does support the early-2002 turnaround.
