StL Fed: 2009 Stimulus --> trash.

Just wanted to share some boring empirical work FWIW. Analysis of the first year shows some... weak results.

Title of the analysis is appropriately straightforward- "The 2009 Recovery Act: Directly Created and Saved Jobs Were Primarily in Government."

The direct numbers look so low compared to initial projections that even the low-end (1 million - 3 million jobs) looks unlikely. They were revised way down, as is mentioned below. This is for the first year of the program- stuff on more recent data will be coming out eventually.

-1.3 - 3.3 million jobs created/ saved was projected. Direct result was a great deal lower, and 75%+ of them were government jobs:

quote:

---

166,000 of the 682,000 jobs directly created/saved were in the private sector. Examples of private sector stimulus jobs include social workers hired by nonprofit groups to assist families, mechanics to repair buses for public transportation, and construction workers to repave highways. Examples of government stimulus jobs include public school teachers, civil servants employed at state agencies, and police officers. While fewer than one of four stimulus jobs were in the private sector, more than seven of nine jobs in the U.S. economy overall reside in the private sector. Thus, stimulus-funded jobs were heavily tilted toward government.

---

-They also survey other recent empirical work, it appears that at best the program generated a small bump at colossal cost, or worse, nothing at all:

quote:

---

while the stimulus was was unsuccessful increating/saving private sector jobs, it helped maintain and sometimes increase (i)state and local government services, and in turn public sector jobs

---

quote:

---

Conley and Dupor (2013) use state-level data to conduct a cross-sectional study of the
effect of stimulus jobs. They find that the Act resulted in a statistically significant increase in state and local government employment but not in private employment.

---

quote:

---

Jones and Rothschild (2011), using their own surveys of grant, loan, and contract recipients, find that approximately one-half of the individuals filling positions directly created by Recovery Act funding were leaving other jobs

---

Oof!

quote:

---

Wilson (2012) uses state-level variation through an instrumental variables method to
study the job effects of Recovery Act spending. His results concerning the private sector effect
of the stimulus are mixed

---

(That "mixed" doesn't mean that some of them were actually negative. It means that some all were positive, but only some were statistically significant.)

quote:

---

Michael Grabell’s (2012) book, Money Well Spent? The Truth Behind the Trillion-Dollar
Stimulus, the Biggest Economic Recovery Plan in History, intended for a general audience, gives
an excellent account of many aspects of the Recovery Act. It contains a detailed narrative
account of why the design and execution of the stimulus led to slow and limited direct job
creating/saving in the private sector, such as transportation, but rapid job creating/saving in
the government sector

---

quote:

---

Several times per year since the beginning of the stimulus plan, the CBO has published a
low-high interval estimate of the employment effects (combining direct and indirect jobs
created/saved) of the program. In its first nine reports, the CBO projected that between 1.3
and 3.3 million persons would be employed in 2010 as a result of the Act; however, in its 10th
report (November 2011), the CBO revised its estimate and reported that the Act may have
created/saved as few as 650,000 jobs in 2010

---

quote:

---

Ramey (2013), using several different specifications of structural vector autoregressions, shows that in response to an increase in government purchases, government employment rises while private sector employment falls or is unchanged

---

LINK