How much will it cost to fence up the border?

Last Wednesday, Felona asked “How much will it cost to fence up the border?”  That’s a great question.  First of all, there aren’t any perfect answers.  Until it is actually “fenced up” we can only estimate, but I will give you the best estimates I can find.  Before diving into the numbers, I must say I was shocked at the price tag.  I guess when I hear the word “fence” I think of the chain link fence in my backyard, or the wood picket fence my grandparents had.  But the “Border fence” is actually more like a prison wall, or even multiple prison walls like the walls in the designs below.

Two Walls with barbed wire and ditches with motion detectors concept picture

A concept design of a Border wall(s)

Picture of multiple walls, cameras, and motion detectors

Actual Image of a Border Wall(s)










So, the question remains, how much will these walls cost?  Let’s try one part.  Boeing was payed 67 million dollars to create a 28 mile “virtual fence”.  Based on the fact that the U.S. Mexico border is 1993.4 miles(pdf), it comes down to a math problem.  67 million/28 miles * 1993.4 miles = nearly 4.8 billion dollars to build.  Unfortunately, it doesn’t end there.  Due to delays, glitches, and other cost overruns, The fence actually ended up costing  almost a billion dollars to span just 53 miles.  Plug that into your calculator and you get 37.6 billion dollars.  A lot higher than earlier reports of only 6 or 7 billion.

Of course, other, less costly style border walls have gone up.  But even those will be upgraded with additional fencing and equipment.  The problem is that a single fence is not good enough.  A single fence can be scaled, dug under, etc…  It requires something elaborate to actually stop people.  So I think it’s fair to estimate the cost of a Boeing style wall:  Double walls, specially built with barbed wire or slants at the top to make it difficult to climb over, separated by cameras and other sensors to alert border agents.

To check my numbers, I found this congressional research service report from a couple of years ago about the status of the border wall.  It threw out a lot of estimates on per mile costs of the fence:

The Congressional Budget Office (CBO) has estimated that border fencing would cost $3 million
a mile to construct and that maintenance would total roughly 15% of the overall project costs per
year.103 However, the CBO does not elaborate on what is included in those estimates. DHS
predicts that the San Diego fence will have a total cost of $127 million for its 14-mile length
when it is completed—roughly $9 million a mile. Construction of the first 9.5 miles of fencing
cost $31 million, or roughly $3 million a mile, while construction of the last 4.5 miles of fencing
is projected to cost $96 million, or roughly $21 million a mile.104 However these costs may be
somewhat misleading due to the following factors: construction of the fence was delayed for an
extended period of time; the remaining construction involves filling a relatively large gulch which
may be more complex than the average stretch of border; and DHS is proposing to use private
contractors to expedite the construction process which may increase the labor costs and thus may
increase the overall project costs. According to the Government Accountability Office (GAO), the
border fencing constructed by the end of FY2007 (using mostly the Corps of Engineers and the
National Guard to construct the fencing) cost about $2.8 million a mile. The fencing constructed
in FY2008, using mostly private constructors, cost about $5.1 million a mile.105

All those estimates are between 3 and 9 million dollars per mile(I left out the 21 million dollar per mile estimate since, as the report described was an unusual geography).  That’s between 6 and 18 billion dollars to complete.

Of course, that is just the cost to complete the fence.  That doesn’t cover maintenance of the border fence.  There is a very real possibility of people knocking down portions of the wall or nature just washing away 40 feet of it at a time.  In the same report at the one above, the army corp of engineers estimated the maintenance costs to be between 16.4 and 70 million dollars per mile over a 25 year period.  That comes out to 650 thousand and 14 million per year per mile.  Or put another way, between 1.3 billion and 5.6 billion per year just to maintain and repair the wall after it is completed.

Not a direct cost, but the Sierra Club and others have noted the damage to wildlife that a border fence would pose.  It is unclear to me how much that would cost business or government to either solve or to endure it.

In conclusion, I have to say while building and maintaining a border wall along the U.S.-Mexico border is very expensive, it is not completely impractical.  Even taken the highest estimates to build and maintain, we are talking about between 1 and 2% of the average Federal Budget to build the fence and less than 1% per year to keep up.  I’m happy to offer some the real, concrete numbers. I’ll let others argue about whether it is worth the costs.  I think my position has been made clear in the past.  Thanks for the great question, Felona!

CATO’s Griswald Agrees with my Immigration conclusion

Last month I wrote a post in which I showed the number of undocumented workers vs. how much we spend on border security.  My conclusion was that the amount spent on border security had little effect on immigration.  The numbers are the best evidence:


Daniel Griswald over at CATO’s liberty blog comes to the same conclusion in his post, What’s Behind the Decline in Immigration?  It’s the economy, Stupid. He writes:

The more obvious explanation is the steep economic recession that began to bite in 2008. The downturn has been especially brutal in the housing and construction industries where many illegal immigrants found employment during the previous boom. As evidence, the decline in the number of illegal immigrants has been steepest in those states, such as Nevada, California, and Florida, where the housing downturn has been the most severe.

Libertarians.  What can I say, we disagree on macroeconomics, but can agree on this and the need for comprehensive immigration reform.

Border Secuirty and Jobs Bills Pass

As I predicted last friday, the house has passed the “let’s flush $600 million down the toilet in the name of border security” bill.  I won’t rehash why the money is being wasted.  From Reuters:

The U.S. Congress on Thursday passed legislation to strengthen security along the border with Mexico, trying to tackle the politically sensitive issue of illegal immigrants ahead of November congressional elections.

Final legislative action came as the Senate passed the bill on a voice vote, one day after the House of Representatives interrupted a six-week recess to pass the bill.

The only good thing to come out of the bill is it’s funding mechanism.  In order not to add to the deficit congress had to either cut spending or increase revenue.  In this case they mostly chose to raise revenue.  However, instead of a general tax increase they targeted towards companies that make a business model out of importing cheap labor into the company.

The plan is financed by imposing higher visa fees on some Indian technology companies operating in the United States, prompting a protest from the government of India.

Senate aides said the Indian companies had been targeted because they take advantage of a U.S. law to import a high percentage of their workers from abroad. They said four Indian companies would be affected: Tata (TCS.BO), Infosys (INFY.BO), Wipro (WIPR.BO) and Mahindra Satyam (SATY.BO).

It may strange that congress would pass a bill that hits only 4 companies.  However, that’s because they are the only 4 large companies whose majority of state-side employees hold temporary visas.  Fee qualification is explained over at NDTV:

The Bill proposes to increase visa application fees by at least USD 2,000 for the next five years to raise nearly $550 million to help fund the $650 million plan for increasing security along the US-Mexico border. These fee increases will apply only to companies with more than 50 employees and for whom the majority of their workforce are visa-holding foreign workers.

Indian software firms, including IT biggies like TCS, Infosys, Wipro and others, use H-1B and L-1 visas to fly their employees to the US for working at their clients’ locations as on-site engineers.

The bigger news is that congress also passed a so-called “jobs” bill.

House Democrats on Tuesday pushed through a $26 billion jobs bill to protect 300,000 teachers and other nonfederal government workers from election-year layoffs, while sending $16billion to help cash-strapped states with rising Medicaid costs.

The bill would be paid for mainly by closing a tax loophole used by multinational corporations and by reducing food stamp benefits for the poor. It passed mainly along party lines by a vote of 247-161.

I’m all about closing tax loopholes, and I’m also glad that we’re going to keep more teachers in the classroom, and police on the streets, but I am a bit concerned about cutting food stamps during the worse recession in 70 years.  Cutting the food stamps keeps the bill “budget neutral”.  However, it’s likely just smoke and mirrors anyways.  The food stamp funding will likely be restored in next years budget.  It’s hard to vote against giving food to the poor.

Under the bill, effective March 31, 2014, food stamp benefits would return to the levels that individuals would have received under pre-Recovery Act law. This modification is estimated to save $11.9 billion over 10 years. House Democrats plan to work to restore this funding before the cuts are implemented in 2014, however.

President Obama signed the Teacher and Police jobs bill into law yesterday and will be signing the border security bill into law today.  All this when the congress is supposedly on their “august break”.

Indian software firms, including IT biggies like TCS, Infosys, Wipro and others, avail H-1B and L-1 visas to fly their employees to the US for working at their clients’ locations as on-site engineers.

Another $600 million wasted on Border Patrol

I didn’t realize that my Wednesday post, Increased Border Security Doesn’t Stop Immigration, would end up being so timely.  Last night, the Senate voted to spend another $600 million on immigration enforcement.  From the AFP.

The US Senate has voted to beef up US-Mexico border security with another 1,500 agents and more unmanned aerial vehicles that scan the frontier for undocumented immigrants or drug runners.

The legislation’s 600-million-dollar price tag would be paid for by raising fees on what the measure’s backers called a handful of foreign firms that exploit US visa programs to improperly import workers to the United States.

The measure includes money for 1,000 new US Border Patrol agents to form a “strike force” for quick deployment, 250 new Immigration and Customs Enforcement agents as well as 250 new Customs and Border Protection officers at ports of entry, and to boost communications among law-enforcement officials.

What’s galling about this is that all this money being spent is essentially being thrown away.   As I showed in my last post, in the past, no matter how much we spend on past enforcement techniques, the number of illegal immigrants never went down.  Maybe this money wouldn’t be a waste if they were going to try something new, but there is no new program here.  I seriously doubt that the impact of this will be anything but a small blip – if that.

(Click chart for larger image. Click here for the numbers)

Another point that suggests this is a waste is where the money is going.  45% of immigrants who aren’t here legally are from expired visas.  However, of the 1500 new enforcement agents that are being hired, 67% of them are going to patrol the border, 17% are going to other ports of entry, and only 17% will be in Immigration and Customs Enforcement – not all of which are guaranteed to be immigration enforcers.

This bill will likely become law.  The bill passed on a voice vote so we know that it’s a pretty popular bill in the senate.  Now that it’s passed its going to go to the house of representatives.  Considering a very similar bill (costing $701 million) had already passed the house of representatives (239-182), that this Senate bill will easily pass.  The only problem is that the house is currently adjourned for the august recess.  However, it will be called back into session to pass another bill that the Senate has sent them.  Perhaps they will pass this bill while they’re at it.

Increased Border Security Doesn’t Stop Immigration

Immigration reform is a hot topic right now.  There’s a lot of calls for more border security and increased enforcement.  However, do these policies work to actually reduce the number of illegal immigrants in the country.  Every year we spend more on enforcement and almost almost every year the number of undocumented workers living in the country still  grows.

I wanted to create a graph that showed the number of undocumented workers in the country vs. the amount we spend on immigration enforcementborder security over time.  The problem is that there is surprisingly little data that estimates the number of illegible immigrantsundocumented workers living in the country.  The Department of Homeland Security created official estimates(pdf) for the years 2000, 2006, 2007, 2008, and 2009.  The rest of the years I would have to find other sources.  Fortunately, a website called procon already did that.  They compiled a list of illegal immigrant estimates from several different sources.  The data goes back to as early as 1969, but doesn’t have separate numbers for each year.  However, I think that’s the best we can do so I’ll have to make do.

Here is the graph of the amount we spend on immigration enforcement vs. number of undocumented workers living in our country.  The amount spent has been adjusted for inflation and made into 2009 dollars.

(Click chart for larger image. Click here for the numbers)

As you can see the amount we spend on enforcement – even adjusting for inflation – has been going up since the mid 80’s.  However, the number of immigrants keeps going up.  It is not until the enforcement spending skyrocketed in 2008 and 2009 that the number of immigrants actually decreased.  That is the only time there has been a correlation between increased border patrol spending and a decrease in illegal immigrants.  The rest of the time there is no correlation between increased spending and falling undocumented immigrants.  In fact, the numbers almost seem to rise and fall at the same time.

Some might argue that we’re finally spending enough on border control and immigration enforcement that we can finally see the results.  However, let’s take a look at the graph above.  This time, however, I’ve added in the United States GDP (also adjusted for inflation) into the mix.  (Border Patrol spending is in thousands of dollars, GDP is in millions of dollars)

(Click chart for larger image. Click here for the numbers)

Looking at this graph we can see 3 instances where a decrease or a flatline of U.S. gdp(i.e. a recession) has been a leading indicator to fewer undocumented workers.  If there weren’t so many gaps in the historical data of illegal immigrant populations, we’d probably see even more.

I think the case for an enforcement only immigration policy is pretty thin for two reason. One, only once has increased spending lead to fewer immigrants, and that may yet still prove to be because of the recession.  Two, look at how much we’ve increased spending on enforcement since 1962.  We’ve been using an enforcement only policy for years and do you think  its working?

Finally, one last interesting statistic.   The federal government spends more on border patrol and immigration enforcement than the rest of it’s law enforcement activities combined.  In 2009 we spent  $27.5 billion total on law enforcement activities.  $17.2 billion of that was spend on border patrolcustoms and immigration enforcement(see numbers here).  So the FBI, Secret Service, DEA, ATF, and U.S. Marshals all combined cost less than our current immigration enforcement policies.

There must be a better solution for dealing with undocumented workers than to spend 62% of all federal law enforcement.  Based on these numbers I find it hard to see how an enforcement only policy can work.