The Case Study on SEMPRA’s Outsourcing Plan of Green Jobs to Mexico

by on September 7, 2011 · 2 comments

in California, Energy, Environment, Labor

Editor: The following is an edited “executive summary” of Peter Philip’s case study – published June 10, 2011 – on SEMPRA’s plan to outsource an estimated 15,000 jobs to Mexico by building a cross-border 1250 megawatt energy transmission line to their energy facility in Mexico.  The full report can be found here.

Should Green Jobs Be Outsourced? A Case Study of Lost Jobs and Lost Opportunities

By Peter Philips, PhD

The proposed Sempra 1250 megawatt (MW) tieline connecting the California grid to envisioned new wind-farms in Mexico is not just about electricity. It is also about foregone opportunities, lost human capital investment, lost worklives, lost tax revenues, and diminished economic development prospects; and also, it is about which regulatory authority, California or Mexico, should oversee the environmental impacts of building green generation capacity for the California grid. Finally, it is about undoing some of the economic benefits and jobs stimulated by the first set of federally subsidized, utility-scale, solar projects fast-tracked by the Interior Department.

Approving the Sempra tieline into Mexico will result in:

  • 5 years of lost construction work
  • 3000 lost construction job-years including 2450 lost to Imperial County, California residents
  • (At 27.9%, Imperial County has the highest unemployment rate in the country)
  • 3450 lost job-years overall in Imperial County (2450 construction plus 1000 spinoff jobs)
  • 9800 lost job-years in California, including Imperial County
  • 15,000 lost job-years in the US. including California
  • $550 million in lost wages (plus additional losses in benefits) over 5 years
  • $300 million loss in local, state and federal business and personal taxes
  • $4.5 million loss in local human capital investment in Imperial County
  • 103 Imperial County youth deprived of apprenticeship training and skill acquisition
  • $127 million net present value of lost lifetime wages and benefits associated with foregone training
  • 40 permanent operation and maintenance jobs lost in Imperial County amounting to 1000 lost job-years over 25 years
  • $78 million in the net present value of lost wages and benefits in Imperial County from the lost operations jobs in addition to the aforementioned $550 million in lost wages associated with lost construction and related spinoff jobs

Sempra, the parent company of San Diego Gas & Electric, proposes to outsource 1250 MW of green electrical generating capacity to Mexico by connecting a one-mile tieline from Mexico to the Southwest Powerlink electrical transmission line close to the San Diego County and Imperial County border. (See map) This proposal, if approved, would outsource to Mexico 5 years of work for 600 construction workers, 489 of whom otherwise would have been residents of Imperial County, and 111 of whom would have traveled from other parts of California, Arizona and Southern Nevada to work in Imperial County. In total, 3000 direct, on-site construction job-years (600 workers times 5 years) will be taken from the U.S. construction labor market and transferred to Mexico.

In addition to these lost construction jobs, 40 permanent operation and maintenance jobs will be lost. Over a 25 year lifespan of these types of facilities, 40 lost operations jobs amount to 1000 lost local job-years taken from the Imperial County labor market. Overall, the U.S. labor market would lose 4000 job-years in direct, construction and operation work by approving Sempra’s proposal. Almost 90% of these direct job-losses would be lost in the Imperial County labor market itself.

Imperial County can ill-afford to lose any jobs. In April, 2011, Imperial County’s unemployment rate stood at 27.9%, the highest county unemployment rate in the nation. Construction employment in Imperial County is down 50% from early 2008. The loss of 489 long-lasting construction jobs in this small county of 170,000 people, at this time of deep economic crisis simply rubs grit and salt into an already gaping wound. But the job losses from Sempra’s proposal do not stop at the construction-site gate nor at the county line.

The loss of direct construction employment would lead to a spinoff of an additional 1000 lost job-years elsewhere on other types of jobs in the local Imperial County labor market. Because the Imperial County population is small, the spinoff losses from outsourcing these jobs to Mexico will spread to the overall California and U.S. labor markets through supply-chain and consumer-chain channels. In total, outsourcing this work to Mexico will mean that the U.S. labor market will lose from 10,000 to 15,000 job-years, a multiple of from 3 to 5 times the direct job-years lost on the construction work itself. These 15,000 lost job-years correspond to net present value of $550 million in lost earnings.

With the loss of direct construction work alone, comes more than $4.5 million of lost human capital investment through the loss of more than 100 apprentices whose 5 years of training would have been financed by this construction work. …

Of the 103 apprentices that would have been on this Imperial County work, 75 would have been electrician apprentices. Each of the 75 skipped-over electrical apprentices regrettably will forego more than $36,000 of human capital investment in classroom and lab training that contractors otherwise would have invested in them. This lost human capital investment is equivalent to what the State of California invests for the first three years of a student’s undergraduate training at the University of California. Outsourcing this type of work to Mexico is like closing a university in the harm that it does to post-secondary education for blue-collar workers in Imperial County.

The loss to these individuals is also a loss to Imperial County which will lose the services and economic development advantages of having 103 additional, highly skilled construction workers within the local construction labor force. Given that the total size of Imperial County’s construction labor force is about 2000 workers, this amount to a 5% loss in the total skill makeup available to local construction contractors. …

The loss of construction jobs and consequent loss of spinoff jobs during the period of construction alone reduces local, state and federal tax revenues by almost $300 million. More lost tax revenues are associated with the absence of these power-generating facilities within the County over the 25-year, expected lifetime of this power generation. Sempra’s tieline proposal is about importing electricity, but it is also about outsourcing jobs, foregoing human capital investments, lost careers and lost tax revenues.

FAQs: Concepts and Conclusions

These frequently asked questions will help the reader understand some of the conceptual issues and conclusions found in this report.

What is a “job” in analyzing job losses?

Whenever someone gets hired, that person has a new job. But that new job could last 4 months, 4 years or 40 years. Obviously, there is a lot more work, income and spinoff effects from 40 years of work compared to just 4 months of work. So in analyzing job losses or job gains, economists have standardized the concept of “job” as a “job-year.” A job-year is 52 weeks of 40 hours of work per week, or 2080 hours equaling one year’s worth of work. When an economist looking at the economic impact of building a new solar farm says that this solar farm will create 3000 new jobs on the construction site, that economist means 3000 new job-years.

Does that mean there will 3000 construction workers on the construction site?

Not necessarily. If a construction project requires 3000 job-years, and the construction will last one year, than we would expect, on average, 3000 construction workers on that worksite. However, if the job is expected to take 5 years, then we would expect, on average, there would be 600 construction workers on the worksite at any given time. (600 workers time 5 years equals 3000 job-years).

What are the spinoff or multiplier effects of 3000 new jobs from a new construction site?

Any construction site requires materials as well as workers. The materials bought for the new construction site will create new jobs somewhere else in order to make and transport these new construction materials to where the new construction is taking place. This new upstream demand will create new jobs spun-off from the new construction work. Also, the workers on the new construction site will spend their new wages buying food, paying for their homes, buying cars, gas and other consumer goods and services. This new consumer demand will create new spinoff jobs downstream in the consumer market.

If a new green power-plant is built in Mexico instead of in the U.S., wouldn’t the American construction workers just go work somewhere else?

Sure, if there was full employment. But today we have the worst labor market since World War Two. California is one of the hardest hit states while Imperial County, with an unemployment rate of almost 28% in April 2011, is the hardest hit county in the nation. Furthermore, construction employment in Imperial County is only half of what it was in early 2008. So a job lost to Mexico in the aftermath of the Great Recession means an unemployed American construction worker stays unemployed.

How do you calculate the multiple number of spinoff jobs from a new construction site?

The new construction jobs are called the “direct” employment effect of the new work. The new spinoff upstream supply-chain jobs are called the “indirect” employment effect. The new spinoff downstream consumer-chain jobs are called the “induced” employment effect. The “multiplier” effect is the multiple number of new indirect and induced spinoff jobs that are created by the original new direct construction jobs. So the total number of new jobs is the direct jobs plus the indirect jobs plus the induced jobs. The multiplier is the total number of jobs divided by the original new direct construction jobs.

The multiplier effect from the original number of new direct construction jobs depends on how big an area you are looking at. In a small county such as Imperial County, the multiplier will be small because the upstream supply-chain and the downstream consumer-chain will both be short. The solar panels built for a solar farm will not be built in Imperial County. So any new jobs created by a demand for solar panels will not create those new jobs in Imperial County. …

But as you lift your gaze from Imperial County to California or higher still to the U.S. economy as a whole, the supply-chains and the consumer-chains will get longer; and the potential for new spinoff jobs making the construction materials or making the consumer goods to meet this new demand will grow substantially. …

How many jobs in total would be lost in the U.S. if Sempra is allowed to build a 1250 MW transmission line and import green energy from Mexico?

We calculate that 3000 direct construction jobs would be lost with 2445 of those being lost by Imperial County construction workers. These 3000 jobs are measured in job-years, so if it took 5 years to put in place 1250 MW of photovoltaic generating capacity, then 600 individual construction workers would lose 5 years worth of work each while 489 of those individuals would be Imperial County residents.

In addition to these lost construction jobs, there would be almost 400 supply chain jobs and 600 consumer chain jobs (measured in job-years) lost in Imperial County for a total Imperial County loss of 3439 jobs (again measured in job-years). But California (counting Imperial County) would lose more. The loss of the original 3000 construction jobs would lead to a total loss of 9787 California jobs with more new jobs lost in the consumer-chain than in the supply-chain.

But the U.S. (counting California) would lose the most. The original 3000 lost construction jobs would lead to almost 15,000 lost new jobs overall with half of that overall job loss coming from lost consumer demand, a 30% from lost producer demand and 20% from the lost direct construction jobs themselves. These 15,000 lost job years imply $550 Million in lost earnings.

What would be the tax loss associated with these lost jobs?

At the federal, state and local levels taken together, the tax revenue loss would be almost $300 million.

Construction workers do not stay on one job indefinitely. Are there any long-term losses that continue past the lost jobs on the actual construction work itself?

Yes. Had Sempra built its green generation capacity in Imperial County instead of Mexico, contractors on the U.S. side of the border would have invested more than $4.5 million over 5 years in classroom and lab instruction for the more than 100 apprentices that would have worked on this job.

For electricians, the largest group of apprentices, contractors would have invested more than $36,000 over 5 years in classroom and lab instruction, plus provided hands-on, supervised on-the-job training. This lost human capital investment means that these 103 would-be apprentices will forego skills development, get less well-paying work, and each earn, a net present value in today’s dollars, almost $1 million less over their worklives than they would have earned had they received this training.

To give you an idea of the value of this lost training, the foregone human capital investment due to this lost work is equivalent to about what the State of California invests in the education of a University of California undergraduate over his or her first three years.

Outsourcing a large, long-lasting construction site is like closing down a university. Indeed, construction apprenticeship training is the largest system of privately financed higher education in the United States creating well-paid, middle-class, blue-collar jobs by investing huge sums in human capital on local youth while essentially putting each apprentice on scholarship because each apprentice earns while he learns. To make the apprenticeship system work, contractor-paid-for apprenticeship programs rely upon projects like the type Sempra proposes to outsource to Mexico. Imperial County also loses because local communities rely upon human capital investment in their young people to build a key component of future local economic development, namely a skilled local construction labor force.

You say that Sempra intends to build wind farms in Mexico, yet you use as your alternative the building of solar farms in Imperial County. Why?

Currently a megawatt of wind energy is cheaper than a megawatt of solar energy, although solar energy costs have been falling. The places where wind farms can be built are relatively limited and probably all of California’s wind resources that can be developed responsibly with due consideration for environmental impacts will, in fact, be developed whether or not Sempra is allowed to construct its tieline.

California requires that all utilities and other electricity providers in California get 33% of their electricity from renewable energy sources such as geothermal, wind and solar by the year 2020. To achieve this goal, all of the geothermal and all of the wind resources in the state that can be developed responsibly with appropriate consideration of environmental impacts will be developed; and still the 33% renewable-energy generation standard will not be fully met. So to build up to this 33% goal, utilities will have to develop solar resources as well. Thus, at the margin, if 1250 MW of wind energy is not developed for the California grid in Mexico, then 1250 MW of solar capacity will be built in California. While the construction of thermal solar-farms are breaking ground now, the recent decline in the cost of photovoltaic solar generation will mean that future solar farms will likely be photovoltaic.

You point out that wind energy is cheaper than solar energy. Isn’t it better to build across the border in Mexico to capture this wind resource in order to benefit from the cheaper cost of wind?

Picking up an additional large wind farm in Mexico would probably lower San Diego Gas & Electric’s green electrical generation costs somewhat. And these savings would mostly be passed on to at least some SDG&E customers.

However, the U.S., and especially California, carefully evaluates the potential environmental harm of building any type of power plant–gas, solar or wind. Mexico does not have as careful an environmental review. So the cheaper wind-generated electricity built in Mexico may have hidden environmental costs that could make that imported wind-energy cost artificially low.

In any case, Mexico needs its own green-energy generation capacity , including wind and solar. Mexico disproportionately relies upon high-sulfur-content oil for much of its electrical generation This form of power production generate more pollution with consequent increased health hazards. By building a captive wind-farm in Mexico tied to the California grid, Mexico is deprived of this wind-resource. So another hidden cost of the proposed “cheaper” wind farm is more pollution and more health hazards in Mexico.

So, while there may be some cost savings to SDG&E associated with capturing a Mexican wind resource for Sempra’s sole use, this savings has to be balanced against the potential environmental, pollution and health hazards Sempra’s proposal entails for Mexico along with the costs in lost American jobs, lost domestic training, lost local income and lost U.S. tax revenues outlined in this report. So, another hidden cost of the proposed “cheaper” wind farm is more pollution and more health hazards in Mexico.

About the Author

Peter Philips is a labor economist specializing in the construction labor market. He is the nation’s recognized expert on the economics of prevailing wage laws and one of the foremost experts on the construction labor market, generally. Philips has related interests in construction worker safety, health economics and economic history. Philips received his B.A. from Pomona College and his M.A. and Ph.D. from Stanford University. He is Professor of Economics and former chair of the Department of Economics at the University of Utah.

Philips’ most recent books include Building Chaos: An International Comparison of the Effects of Deregulation on the Construction, (with Gerhard Bosch, 2003) and The Economics of Prevailing Wage Laws, (with Hamid Azari-Rad and Mark Prus, 2005).  Philips has received many awards for his teaching and research including the prestigious University of Utah Presidential Teaching Scholar, and the University of Utah, Graduate Student & Postdoctoral Scholar Distinguished Mentor.

Acknowledgement, Disclosure and Disclaimer

Support for the research for this report was provided, in part, by the California Construction Labor Management Trust. The author has sole control of and responsibility for the findings and views expressed in this report. This report is copyrighted by the author. The author who may be contacted at


{ 2 comments… read them below or add one }

Frank Gormlie September 19, 2011 at 12:28 pm

Whew! Every progressive in San Diego and California needs to read this critique of Sempra’s plans to outsource green jobs to Mexico.


Patty Jones September 19, 2011 at 2:47 pm

One very interesting point for me stems from the fact that since there are limited number of places where wind farms are practical, building this wind farm in Mexico to service SDG&E customers would deprive Mexico (who “disproportionately relies upon high-sulfur-content oil for much of its electrical generation”) of a wind resource in this area.

There is so much talk of our dependence on oil… it doesn’t make sense to me to further another country’s dependence on oil to lessen ours.


Leave a Comment

Older Article:

Newer Article: