Five percent of the U.S. workforce is made up of illegal immigrants, according to the Pew Research Center. If Trump follows through with his mass deportation plan, how, if at all, will these jobs be replaced?

Basic supply and demand theory tells us that in the short run, employers would be scrambling to fill those jobs and would have to increase wages to attract native workers. Given the relatively tight labor market in the U.S., with unemployment under 5 percent, it would be difficult to fill so many jobs all at once. In addition, it’s not clear that native workers would be willing to take those jobs without a significant hike in wages—particularly in jobs in agriculture, construction, and janitorial services where employers can pay illegal immigrants below market wages.

In the long run, it’s unlikely that we would have to replace each and every one of those jobs previously held by someone who was deported. This is because over time firms have more flexibility to change production processes to use less labor and more capital. We’ve already seen an increase in automation in both the manufacturing and agriculture sectors—a trend that we would expect to accelerate if labor suddenly became more expensive with the deportation of so many workers.

In fact, this is exactly what happened during the 1960s when President Kennedy ended the bracero program. The program, which allowed almost half a million people a year to take seasonal work on America’s farms, was said to be driving down wages and taking jobs from Americans. After the program was abruptly ended, there was a small but temporary increase in native farm employment in states where farmers had relied heavily on foreign labor—such as California and Texas. Yet within a few years agricultural employment was again declining similar to states where there had been no braceros. Rather than leading to higher wages and more work for Americans in the fields, farmers turned to machines.

Still, the scale of the proposed deportation is such that it’s likely the U.S. economy will suffer some employment loss and see rising wages in those sectors where illegal immigrants are most likely to work. However, the magnitude of the effects on the labor market will depend on how willing natives are to fill those jobs vacated by immigrants and the degree to which employers can substitute capital for labor.

Shrinking the labor force by 5 percent would likely increase wages and the cost of goods, creating inflation and higher interest rates. What other unintended consequences might Trump’s deportation plan have on the economy both here and abroad?

Few economists would argue that immigration helps fuel the U.S. economy. When immigrants enter the labor force, they increase the productive capacity of the economy and raise GDP. Their incomes rise, but so do those of natives—particularly business owners who benefit from the cheap labor. While empirical estimates peg this “immigration surplus” to be a rather small share of additional GDP—typically 0.2 to 0.4 percent—it still amounts to $36 billion to $72 billion per year. Trump’s deportation plan would certainly mean the loss of some portion of that surplus, leading to slower economic growth.

Immigrants also “grease the wheels” of the labor market by flowing into industries and areas where there is a relative need for workers—they typically move to where the jobs are. This added flexibility is important to help alleviate shortages or surpluses that would otherwise limit economic growth or hinder an economic recovery. For example, during and after World War II, Mexican immigrants were instrumental in alleviating shortages arising from the war effort. Deporting a sizeable share of the most adaptable part of our workforce means that the U.S. economy is less able to adjust to future economic shocks—whether they are booms or busts—as well as demographic changes such as an aging population.

Finally, it’s hard to say how much of a “chilling” effect the deportation plan and the recent travel ban will have on legal immigration. Why is this important? Recent immigrants are much more highly educated than previous generations—among immigrants ages 25 and older residing in the U.S. in 2015, 48 percent of those who arrived after 2010 have a bachelor’s degree compared to only 27 percent of those who arrived in 2005 or earlier. By comparison, 31 percent of native-born adults have a bachelor’s degree.

Recent immigrants are most likely to hold professional or technical jobs such as medical scientists, software developers, physical scientists, and economists. This rise in high-skilled immigration has also been linked to innovation, specifically to higher patenting rates among immigrants.

Thus the recent shift in immigration policy may have far-reaching consequences affecting the ability of the U.S. to attract and retain skilled immigrants on which states such as Massachusetts rely quite heavily to fuel growth in the state’s higher education programs as well as its overall labor force. Moreover, employers are concerned that President Trump will also move to limit the H1B visa program for skilled immigrants, which he has called a “cheap labor program” in the past. As a result, these skilled individuals may be more likely to seek opportunities across the globe as well as in their home countries spurring growth and innovation abroad.

Some states, including Arizona and Georgia, have seen strict immigration laws backfire, leading to hundreds of millions of dollars in losses. What might the Trump administration learn by studying these cases?

Arizona is the poster child of strict immigration laws beginning with the passage of the Legal Arizona Workers Act that tried to regulate unauthorized workers out of the market using E-Verify, an electronic employment eligibility verification system used to weed out unauthorized immigrants when they apply for a job. In the first four years after the law was passed, employment dropped by 15.6 percent in crop production compared to an increase in neighboring New Mexico and California over the same time period. A recent report by the right-leaning Cato Institute concludes that the LAWA explains much of that difference.

Two years later Arizona passed SB 1070 to enforce immigration laws outside of the workplace. The anti-immigration laws exacerbated the impact of the Great Recession with 200,000 forced out of the state and others deterred from entering. Estimates suggest that the drastic reduction in the supply of illegal immigrants in Arizona has reduced the state’s GDP by an average of 2 percent per year between 2008 and 2015.

Keep in mind that the impact of reduced immigration on an individual state such as Arizona is likely to be larger than what one would expect nationwide. This is because immigrants can migrate to other states with fewer restrictions and a more welcoming labor market. However, the lesson is clear that the Trump administration cannot repeal the laws of economics: there will be real consequences for both the labor market and the economy if the U.S. follows through on a mass deportation of such magnitude.

The implementation of Trump’s deportation plan could disproportionately impact the industries in which a large number of undocumented immigrants work, including hospitality, construction, and manufacturing. What kind of ripple effect might potential losses in the hardest hit sectors have on the economy as a whole?

Given that immigration comes with both benefits and costs, it’s important to keep in mind that the proposed deportation plan would entail much of the same but in reverse. In terms of direct effects, we would expect that in the short run, the wages of native workers in the affected industries would increase. However, the higher cost of labor would cause employers to possibly curtail any plans for growth and-or substitute labor for machines, which would ultimately mean lower employment in these industries in the long run. The National Bureau of Economic Research study cited earlier suggests that at an average of $500 billion in output a year, removing all illegal immigrants would be like losing the equivalent of Massachusetts from the U.S. economy.

In terms of indirect effects, the cost of goods in these industries would rise in the short run, meaning that consumers would likely be paying higher prices for all sorts of goods including private household services, fresh fruits and vegetables, and apparel. This would serve to increase inflation and potentially motivate the Federal Reserve to raise interest rates. All in all, we would expect slower economic growth in the wake of a mass deportation—the exact opposite of what the Trump administration is promising in its first term.