Intuition and Economics

(This is the second installment of our series describing how the BLS might respond to critics of the Birth/Death adjustment.  See Part One here.)

Everyone uses intuition, making decisions that seem instinctive rather than the result of a carefully reasoned process.  This type of decision making has actually been the subject of formal study.  For example, Yehezkel Dror's Public Policymaking Reexamined highlighted the use of "extra-rational" processes to make decisions.

The TV version of this is almost a cliche', but we can still learn from it.  The veteran cop and the rookie are on patrol.  The savvy vet senses something wrong.  A few questions make clear that there is a serious criminal behind someone's bland exterior.  There are even some studies supporting the cliche'.

It is noteworthy that those whose intuition is helpful are those most experienced and adept in their business.  A top poker player or bridge player has great "table feel."  Many famous traders are noted for their ability to read the tape.

So here is a question:  Why do people who are not expert in economics believe that they have such great economic intuition?  For most of us, it is better to rely upon data.

Intuition about Job Creation

A case in point is the vise-like grip of misinformation about measuring job creation.  Here is an example relating to the last employment report.  Regular readers of "A Dash" know that we respect the work and the significant influence of John Mauldin.  Many people rely upon his weekly letters.  We were disappointed to read the following:

...Last month saw the number of unemployed rise by 345,000. What was not in the headline data was that 217,000 of those jobs were estimated from the "birth-death" ratio. The US economy creates new businesses that do not get counted in the data, so the BLS estimates what that number is, using previous data patterns. When the economy turns, it overestimates new jobs in recessions and underestimates them in recoveries. No conspiracy, it is just the best methodology we currently have.

But does anyone really think 200,000 jobs were created last month? The real number of lost jobs is worse than the headline. And next month the birth-death number will likely be over 200,000 again.

Mr. Mauldin clearly believes that new jobs are not being created.  He is so confident that he poses it as a rhetorical question.  He shares this intuition with nearly everyone.  It seems obvious.  If labor conditions are tight, there are no new jobs.

Intuition about economics leads people to think in black and white terms, rather than seeing quantitative responses and changes at the margin.  Instead, let us consider actual data, the most recent available from the Business Dynamics series.  It uses the records from state unemployment offices -- no projections, and no one is paying insurance premiums on non-existent jobs.  Here is the summary table of seasonally adjusted data.  (The full year data illustrate the same point).

Table A. Three-month private sector gross job gains and losses,
seasonally adjusted

------------------------------------------------------------------------
     |  3 months ended  
     |-------------------------------------
     | Sept. | Dec. | Mar. | June | Sept. 
  Category  | 2007 | 2007 | 2008 | 2008 | 2008 
     |-------------------------------------
     |  Levels (in thousands) 
----------------------------------|-------------------------------------
     | | | | | 
Gross job gains...................| 7,323| 7,676| 7,130| 7,258| 6,822 
 At expanding establishments.....| 5,849| 6,220| 5,731| 5,858| 5,504 
 At opening establishments.......| 1,474| 1,456| 1,399| 1,400| 1,318 
     | | | | | 
Gross job losses..................| 7,564| 7,366| 7,400| 7,751| 7,754 
 At contracting establishments...| 6,209| 6,010| 6,047| 6,277| 6,383 
 At closing establishments.......| 1,355| 1,356| 1,353| 1,474| 1,371 
     | | | | | 
Net employment change (1).........| -241| 310| -270| -493| -932 
     |-------------------------------------
     |  Rates (percent)  
     |-------------------------------------
Gross job gains...................| 6.4| 6.8| 6.2| 6.4| 6.1 
 At expanding establishments.....| 5.1| 5.5| 5.0| 5.2| 4.9 
 At opening establishments.......| 1.3| 1.3| 1.2| 1.2| 1.2 
     | | | | | 
Gross job losses..................| 6.7| 6.5| 6.5| 6.8| 6.9 
 At contracting establishments...| 5.5| 5.3| 5.3| 5.5| 5.7 
 At closing establishments.......| 1.2| 1.2| 1.2| 1.3| 1.2 
     | | | | | 
Net employment change (1).........| -.3| .3| -.3| -.4| -.8 
------------------------------------------------------------------------
 1 The net employment change is the difference between total gross job
gains and total gross job losses. See the Technical Note for further
information.

{Emphasis added to the line showing job gains at opening establishments}

We can see that in the third quarter of 2008, there were over 1.3 million job gains at new establishments and 5.5 million new jobs at continuing businesses.  The entire table covers data during the current recession.  The results from the 2001 recession were similar.

Briefly put, there is a massive turnover of jobs lost and jobs created every quarter.  The net changes that we see in the news are much smaller by comparison.

Measuring New Jobs

How should the BLS account for the job gains at new establishments?  

Most readers will be surprised to learn that the Birth/Death adjustment is not the principal method.  More importantly, the BLS basically ignores non-responding firms in the business survey.  It knows that some of these are business deaths, but not which nor how many.

The BLS uses a two-step process, described as follows:

Step One - Employment losses from business deaths are excluded from the sample in order to offset the missing employment gains from new business births. Because employment increases from births nearly offset employment decreases from deaths in most months (as illustrated above by the BED data), this step accounts for most of the net of business birth and death employment.

Operationally this is accomplished in the following manner each month. Business deaths that are non-respondents to the survey are automatically excluded because they have no current month data. Death establishments that report zero employment to the survey for the current month are treated the same as non-respondents and also excluded. As a result, the over-the-month change calculation from the sample is based solely on continuing businesses.

For the months subsequent to a business death, the deaths are "kept alive" in the CES estimation process; the growth rate of the continuing units in the sample is applied to them each month. This estimates for the growth of the new business births in the months after their birth but before they can be brought into the sample.

This step accounts for most of the net birth/death employment but not all of it. The residual net employment that is not captured by this step is estimated through an econometric model, described below as Step 2.

Step Two - Modeling for the residual of net/birth death employment change. In this step, the CES adjusts its sample-based estimates for the residual net birth/death employment that step 1 misses. This adjustment is derived from an econometric technique known as Auto Regressive Integrated Moving Average (ARIMA) modeling. ARIMA is a standard econometric modeling technique that is often used to estimate relatively stable series. CES refits the ARIMA models each year, for each basic estimation cell, as part of its annual benchmarking process.

The inputs to the ARIMA model are historical observations of the residual net birth/death employment that is not captured by either the sample or the step 1 imputation described above. These historical observations are derived empirically, from the most recent five years of QCEW historical data.

The Birth/Death adjustment that gets so much attention is Step Two in the process.  The critics mistakenly ignore Step One, even though that is the more significant part of the process.  Step One is also sensitive to economic changes, as noted in this paper:

The imputation part of the procedure is directly related to the current sample and is
therefore sensitive to employment trend shifts and turning points.

Summary

In Part One of this series we showed data proving that the BLS Birth/Death adjustment has improved the monthly estimates of payroll employment changes for every quarter since it has been used.

This article, Part Two, shows the reason behind the result.

• There is massive new job creation at all times, even in recessions.  In bad economic times the job gains are offset by even larger losses.
• The BLS estimates most of the job creation by "keeping alive" some of the business deaths.  This process helps to reflect economic turning points since it follows the results from the rest of the sample.
• The Birth/Death adjustment deals only with the residual new jobs.  It has no separate economic meaning.

The BLS critics are mistaken in looking only at the Birth/Death adjustment, when that is only part of the two-step process --- and the less significant part at that.

In Part Three of this series we will look at a few of the specific claims by critics.

Comments

[Chris]

I know it's late but your work is much appreciated.

[oldprof]

Paul -- Thanks for taking the time to comment. This series has taken an exceptionally long period of research. It is nice to know that some find some value.

Jeff

[PaulinKC]

Outstanding Jeff. This article is a great example of how individuals take mental shortcuts when arriving at an opinion.