Monday, December 28, 2009

2009 Year in Review: The Recovery Has Started

In the first post in this series, I looked at how we got here. As a quick recap, at the end of 2008 the economy was facing a deflationary threat. In response, the government put a stimulus program into place and the Federal Reserve lowered interest rates to near zero. Today's question is how has this paid off?

Let's state with manufacturing.

At the national level, manufacturing is rebounding.

Industrial production bottomed mid-year and has risen four of the last five months.

Capacity utilization also bottomed mid-year and has risen four of the last five months.

The ISM index bottomed last year, but has also been rising.

The Philadelphia Fed's index of manufacturing also bottomed at the beginning of this year and has been rising since.

The Chicago index bottomed mid-year. This index appears to be tracking the overall industrial production numbers a bit more closely.

The New York Fed's manufacturing index started to rise in March of this year. It has fallen the last two months and is currently right at the line between expansion and contraction. This is the only manufacturing index to show a decrease, indicating this is probably an outlier. However, it does bear watching.

In short, the manufacturing sector took a big hit at the end of last year. However, national level industrial production and capacity utilization have been increasing for the last few months. The National association of purchasing managers index has risen since the end of last year. All three major regional indexes are showing expansion although one (the Empire State) has dropped the last two months. In short, the vast majority of indicators are showing expansion.

The ISM services index dropped last month to 48.7M. which indicates contraction. However, the overall trend for this number is still higher, as evidenced by the increase that started at the end of last year. The trend is still up.


Let's break the housing market down into new and existing home sales. All the charts are from Calculated Risk.

New Homes

New homes sales have bottomed. However, they are currently near their bottom. They have increased as strongly as existing home sales.

On an absolute level, the total new home sales inventory is now below the historical norm of approximately 300,000 - 400,000. That is a very good development. In addition,

The months of available supply number is also decreasing and is approaching the historical norm as well.

Existing Home Sales

Existing home sales have been increasing since just after the beginning of the year.

This the total inventory of available homes is still above the historical norm of approximately 2,000,000 - 2,500,000, the overall number is still coming down.

The months of supply at the current sales pace is still above the historical norm of roughly 4-5 months, but this number is coming down as well.

So -- what does all of this information tell us?

New homes have cleared excess inventory. That's a healthy development because it indicates the overall glut of that particular market is gone. But while sales have bottomed, they have not advanced as strongly as existing home sales.

The pace of existing homes sales has increased smartly this year. However, we still have excess inventory along with a high months of inventory at the current sales pace.

The housing market is healing as evidenced by the sales pace of existing homes and the inventory situation in the new homes market. However, the pace of new home sales and inventory situation in the existing home sales market is still troubling and indicates we still have some room to improve.

But, considering where we started the year, this is good progress.

Personal Consumption Expenditures

This is a broader measure of consumer activity and therefore more complete than retail sales.

Overall PCEs are trending up, although the rate of increase is moderate; ideally we'd like to see a stronger upward move. Now let's break these numbers down into their smaller subparts from the the largest sub-part to the smallest.

Expenditures on services comprise about 65% of all PCE sales. However notice that overall purchases on services has plateaued over the last roughly two years. There has been a slight uptick over the last few months, but it is not a strong upward move.

Expenditures on non-durable goods has increased over the last few months, but again, it is not as strong a move as we would like.

Durable goods -- which comprise about 11% of PCEs -- have bottomed. There was an upward spike from the case for clunkers program. However, sales fell from that level, although they have bumped up a bit since then.

Overall, PCEs have clearly bottomed. They are moving higher, although at a slower pace than we would like. This situation most likely represents a retrenchment for the US consumer as he spends less and saves more.


For many people this is the one sticking point about the recovery. While the point is valid, it's also important to remember a few basic economic issues. First, the unemployment rate is a lagging economic indicator. That means it will start to drop after the recovery has started. The reason for this is simple; businesses won't start hiring until they are certain the recovery is in place. That means businesses need to need a few quarters of GDP growth before they start to hire workers.

It's also extremely important to remember the severity of job losses at the end of last year and the beginning of this year. Here is a chart of establishment job losses from the Bureau of Labor Statistics:

There are two important points on this chart. First, note the area that is blocked off from the end of last year and the beginning of this year. During this 5-month period the country lost at least 600,000 jobs/month. That means over a 5 months period the country lost 3 million jobs. To put that in perspective, the total number of jobs created during the last expansion was a little over 8 million. Therefore in that five month period the country lost approximately 36% of jobs created in the expansion of '01-'08. In reality, during the recessions the economy has lost most of the jobs created during the last expansion as shown by this chart from the BLS:

In other words, the pace of job losses has been severe to say the least. That means the recovery will be long and difficult. And it has been. The second point from the chart above of establishment job losses shows that we are almost at the rate of 0 job losses. Put another way, we are near a time when the economy should finally start adding jobs sometime in the next few months.

In addition to the pace of establishment job losses there is also the rate of initial unemployment claims which has been dropping as well. Here is a chart of that number:

The pace of claims -- but the weekly and the 4-week moving average -- have been dropping since the Spring. To put that number in historical context, here is the chart of initial unemployment claims going back 4 years:

Notice the steep drop that has occurred all year in this number for most of the year. This tells us that the pace of job losses is clearly decreasing. However, the unemployment rate is still moving higher:

It's important to remember there are two job surveys -- the household and establishment. The unemployment rate comes from the household survey. While the pace of increases is slowing we have not had a firm drop in the overall rate yet. But -- and as mentioned above -- the unemployment rate is a lagging indicator. This tells us the recovery is still new.

And indeed it is. The third quarter of 2009 was the first quarter when the US saw growth in the preceding 5 quarters. A lot of hay has been made of two points. First, the downward revisions which have occurred. The answer to that is simple: statistics are revised all the time -- even years later. And -- more importantly -- considering where the economy was (four quarters of contraction) growth is a good thing. Secondly,

There was much hay made of the fact that of the 2.2%, 1.45% came from motor vehicle output, and especially that this was from the cash for clunkers program. In response, consider these points. First, a large number of the people complaining about that figure are the same people who argued for the first stimulus and in some cases are arguing for a second stimulus. In other words -- there people will complain no matter what. Secondly, government spending accounts for 20% of GDP growth throughout the economic cycle. Finally, it is standard for the government to use programs to bring an economy out of a recession. A favorite method is increasing the upfront depreciation deduction. However, there is nothing wrong with using standard demand simulating measures as well. So, some type of government programs are standard for getting an economy out of a recession.

So in conclusion,

1.) Manufacturing is making a strong recovery.

2.) Housing is on the mend, although there is still work to be done

3.) The consumer is spending again, although weakly

4.) It appears the economy will start to create jobs over the next few months

5.) The economy grew in the third quarter

In short, the recession is clearly over. However, growth is weak. But, considering where we started the year, we're actually doing pretty well.

Finally, let me add my prediction for the recovery, which I posted at the end of August:

I describe the initial phase of the next expansion the "fits and starts" expansion because not one of the four elements outlined about will lead completely or continually. I think it's far more likely we'll see an increase in consumer spending one quarter followed by increased stimulus spending and an increase in exports the next quarter. In other words, various economic sectors will take the lead one quarter and then fall back. In other words, we'll see fits and starts from the above sectors.

I still see a slow recovery with GDP growing between 1%-2% every quarter. It will not be gangbusters growth by any stretch of the imagination.