Theda Skocpol Responds to Judis

Theda Skocpol responds to John Judis's article on why Trump won ...

John, your piece is an elegant example of a genre of post-election autopsy that works no better, I fear, than those polling models.

You offer speculative interpretations of exit poll responses (known to be problematic data) presented as margins for various voter blocs in an aggregate national election. A lot of creative argument that HRC was a poor candidate because voters did not hear the economic message you wish she had delivered. Two problems: national polls showed that voters said she was better than Trump on plans for the economy. That is a small problem, however, because virtually no real policy discussion occurred in this election. Second, huger problem: HRC actually won the national aggregate election you are imagining in the TPM piece by a whopping 2.5 million or more votes. If America were what you measure here, she would be President-Elect.

The problem is that the United States is a federation that conducts fifty separate winner take all plurality elections for president. There too, she lost by a hair in half a dozen states. But the problem was Trump ran up huge margins in nonmetro rural, small town and some outer-suburban areas. Factory workers, even former ones are few and far between there. Previous work shows that Trump voters are NOT disportionately affected by trade disruptions, factory closings, etc. What is more likely is that these nonmetro areas had organized networks – NRA, Christian Right, some RNC and Koch network/AFP presence – that amplified the right media attacks on HRC nonstop and persuaded many non-college women and some college women in those areas to go for Trump because of the Supreme Court.

You say Trump had no organization. True enough for his own campaign. HRC had the typical well-funded presidential-moment machine, an excellent one. We on the center left seem to treat these presidential machines as organization, and they are, but they are not as effective as longstanding natural organized networks. To get some of those working for him, Trump made deals to get the NRA , Christian right and GOP federated operations on his side. They have real, extensive reach into nonmetro areas. But off the coasts, Democrats no longer have such reach beyond what a presidential campaign does on its own. Public sector and private sector unions have been decimated. And most of the rest of the Democratic-aligned infrastructure is metro based and focused. That infrastructure is also fragmented into hundreds of little issue and identity organizations run by professionals.

HRC’s narrow loss was grounded in this absent non-metro infrastructure – and Dem Party losses in elections overall even more so. Obama overcame that deficit. But he is a once in half century figure. How can anyone blame the HRC campaign for failing to equal Obama’s margins among minorities? No Democrat would have done so. For sure, Bernie would not have done so.

Why do these different analytical approaches (aggregate attitudinal vs. organizational) matter? Because they lead to very different prescriptions for what should be done next. Mine says Democrats have to create sustained organizational reach, not just at election time, stretching beyond metropolitan communities and states. Yours, however, is the conventional wisdom: This type of argument is used to argue that Democrats must “message” better and move left on policy issues to attract an imaginary factory-based white working class. How would that have worked in an election where the media never conveyed any policy substance at all? Even next time, if a Trump type does not take over the media, all that approach would do is take the war to imaginary terrain. Failed HRC messaging about trade, etc. was not the reason Trump won. There are few such voters in non-metro America and none would hear trade pact focused messages plausible in the actual lives. In much of non-metro America, families and marriages are fragile, drug deaths are rampant, churches are the only community institutions, men try to piece together service and construction jobs, low paid, while women do the same and try to raise kids. Democrats and their messages hardly penetrate at all, and they seem directed at worlds these people do not live in. Indeed, Dem messages seem directed at blacks and browns – there is a lot of racial anxiety at work.

You just have to get out and drive around America and listen and look to know this is the world that went for Trump and against HRC (and would have gone against Bernie even more). I analyzed the polls from the primaries, by the way: Bernie’s support was young, liberal whites. especially men. In most states, he did not attract extra working class support at all, outside of cities and university communities.

The key for Democrats is to build outward and look for issues that touch the lives of both urban and non-metro families. HRC made headway. More opportunities will soon arrive, for example if Trump/Ryan really do try to privatize Medicare and remove the huge ObamaCare subsidies that help so many in both urban and non-metro areas.

Best, Theda

John Judis replies ...

I want to make three brief responses to Theda’s response to me:

1. I am aware that Hillary Clinton won the popular vote, but I attach little significance to it. She raised and spent twice the money on the campaign that Trump did and had an incredibly larger amount of people on the ground. She had the support of 90 percent plus of the American ruling class, including the current and several ex-Presidents, the united support of her own party, and all of the major media outlets, except for Fox, which was divided. Trump didn’t have even the U.S. Chamber of Commerce on his side. If Trump had had similar resources – if he had blanketed the West Coast where he lost much of the popular vote – the final tally of the popular vote might well have been in his favor.

2. Trump didn’t just win in rural areas and small towns, but in the suburbs – 50 to 45 percent in the final exit polls – where many of the working and middle class live.

3. If you look at a map of where the U.S. has lost manufacturing jobs since 2000, the two leading places are Michigan and North Carolina, and not far behind are Ohio, Pennsylvania (especially in the western half), and Indiana – all states that Trump won, and in the case of all but Indiana, states that Democrats campaigned in, and had won in the past. They are states where many of the voters blame trade and runaway shops -- two of Trump's big issues -- for the loss of their jobs.


Josh Marshall is editor and publisher of