Everyone should think about which topics they want to raise the salience of and why. pic.twitter.com/O9wutKoSVR
— Matthew Yglesias (@mattyglesias) April 18, 2025
Self-criticism corner: By suggesting it is bad to increase the salience of immigration, I have generated meta-discourse that has only further increased the salience.
— Matthew Yglesias (@mattyglesias) April 18, 2025
No more posts about this from me ever!
Yglesias's argument is as follows: Trump's poll numbers on the economy and trade are really bad, much worse than his overall approval rating, so people who want to make him less popular should keep talking about the economy and trade, and stop talking about immigration, because the numbers show that voters like Trump's abductions and deportations.
But the numbers that lead Yglesias to this conclusion can be read in exactly the opposite way: that we're winning the argument on the economy and trade, yet it's not enough to drive Trump's overall poll numbers lower, so we need to go after Trump's strength on immigration and try to lower that number as well.
I'm not sure we can drag Trump's economic numbers any further down, at least in the near future. Yesterday I noted that many of the Trump-voting independents in a recent New York Times focus group have economic anxiety right now but "trust the plan" -- they assume that President Trump is really, really smart and knows exactly what he's doing, and that all of this will work out in the long run. We can try to tell them that it won't work out, but they'll just assume that we're defending the establishment elitists and Trump knows better because he's the world's greatest dealmaker. And then what can we say? We have a pretty good idea what Trumponomics will do in the long run, but until more time passes, we can't prove we're right.
We can see voters splitting on the short-term and long-term impacts of Trump's economic and trade policies in polls. Here are some numbers from a recent CBS poll (apologies for the muddy images; click to enlarge):
In the short term, 75% of poll respondents think Trump's tariffs will increase prices; only 5% think they'll decrease prices. But in the long term, 30% think they'll decrease prices, while 48% think they'll increase prices. In terms of overall economic impact, 65% of respondents think the tariffs will make the economy worse in the long run, but only 42% think they'll make the economy worse in the long run. Only 8% of respondents expect the economy to be better in the short term, but 34% expect it to be better in the long term.
Even the long-term numbers are kind of lousy for Trump, but they're better than his short-term numbers. We simply can't win over the people who expect the plan to work in the long run, because they think Trump is smarter than we are. Also see this result from a March Wall Street Journal poll:
In this poll, 48% of respondents think Trump's policies will "create economic difficulties with very little benefit," and 35% believe they'll "create some economic difficulties in the short run but economic benefits in the long run" (while 13% think they'll just create benefits). The 13% probably can't be won over under any circumstances, and it's unlikely that we can win over the 35% until the economic leopards are actually eating their faces.
Meanwhile, the data analyst who produced the chart Yglesias reproduces above -- G. Elliott Morris, who succeeded Nate Silver at FiveThirtyEight -- disagrees with Yglesias and believes it's worth going after Trump on immigration.
... the individual components of Trump's immigration agenda are much less popular than the general ideas of securing the border or deporting undocumented immigrants.In a piece titled "Trump's Immigration Agenda Is Not Popular," Morris gives us some specifics:
In polls of Trump's immigration policy, voters generally oppose deporting residents who have been in America for more than ~5 years, deporting people where it would separate children from parents, and deporting people who have not been convicted of crimes other than illegal entry.
I keep reading and hearing that, while Trump's net favorability is negative, his approval ratings on immigration are positive. Thanks to @gelliottmorris.com for looking under the hood on this. Turns out Americans *really* don't like Trump's immigration policies.
— Claire Adida (@claireadida.bsky.social) April 15, 2025 at 11:04 AM
[image or embed]
There are some ambiguities in the polling. In the Wall Street Journal poll I cite above, 55% of respondents approve of "deporting illegal immigrants who are suspected foreign gang members to El Salvador without a court hearing to determine whether they belong to a gang"; 43% disapprove. On the other hand, 58% believe that "Donald Trump must comply with federal court rulings that limit his actions or which he disagrees with," while only 37% believe that Trump should refuse to abide by federal court decisions. So if courts block Trump's renditions and express skepticism about whether he's correctly identified gang members, it seems as if there's room for public opinion to change. Contesting Trump's immigration moves involves contesting his characterizations of the abducted. He's obviously lying about many of them, if not all or nearly all of them. Americans want him to deport immigrants who are bad people, but their support for him could change if they don't believe that's what he's doing, and if they see him not merely taking aggressive action but taking aggressive action in defiance of the courts.
I'll add this: mainstream Democrats like Yglesias want the party to be hyper-cautious and limit its rhetoric to as few issues as possible. They seem to believe that voters can't think about more than one issue at a time. Meanwhile, Trumpworld talks about tariffs and Greenland and DEI and the "Gulf of America" and universities and pro-Palestine foreign nationals at universities and alleged immigrant gang members from Latin America and many other subjects, and voters seem able to process it all. Democrats are hemming themselves in -- something they seem to love doing -- if they continually say they mustn't or can't or shouldn't. They should exercise a certain amount of caution, but much less than Matthew Yglesias recommends.
*****
UPDATE: Look who's not afraid to defend due process for immigrants.
Joe Rogan explains to his audience why due process is important and quotes Ben Franklin — "it is better 100 guilty persons should escape than that one innocent person should suffer.” He argues against shipping people to a prison in El Salvador without trial because we think they’re gang members.
— Dare Obasanjo (@carnage4life.bsky.social) April 20, 2025 at 6:50 AM
[image or embed]