US Politics / Election 2016
Russian Interference

What was Russia's real impact on the 2016 election?

Burgess COMMENTARY

Peter Burgess

What was Russia's real impact on the 2016 election?

It’s a fairly straightforward question. But more than a year later, we are no closer to a definitive answer on the actual impact of Russian intelligence hacking efforts as well as their active measures through RT, Sputnik News, and thousands of Facebook and Twitter ads, bots, and trolls on the 2016 election.

It’s not really a question of whether they made a difference: it’s a question of how big or small that difference ultimately was.

This unknown impact would be added to the appeals made by either candidate, the specific states they visited, and how they managed to resonate with the general populace in the wake of the news cycle. And of course there was also the last-minute release of the Comey letter, which FiveThirtyEight states may have dropped Clinton’s numbers by between 3 to 5 percent. Is it possible that she had already been severely hampered by the endless reams of bad news about internal DNC emails, and then John Podesta?

What difference did it actually make?

As this chart from Real Clear Politics averaging polls during the height of the election season shows, Clinton didn’t actually have a lead the entire time. There were many fluctuations, ending with her final 3-point lead on November 7 of 46.8 percent to Trump’s 43.6 percent.


Screen_Shot_2018-02-01_at_08.58.25.png 2016 Presidential Polls June-November 2016

The first public release of the DNC hack occurred on June 14, at which point Clinton had a 6-point lead of 44 percent to 38 percent.

Russian government hackers penetrated the computer network of the Democratic National Committee and gained access to the entire database of opposition research on GOP presidential candidate Donald Trump, according to committee officials and security experts who responded to the breach.

The intruders so thoroughly compromised the DNC’s system that they also were able to read all email and chat traffic, said DNC officials and the security experts.

The intrusion into the DNC was one of several targeting American political organizations. The networks of presidential candidates Hillary Clinton and Donald Trump were also targeted by Russian spies, as were the computers of some Republican political action committees, U.S. officials said. But details on those cases were not available.

On June 15, 2016, Guccifer 2.0 released the DNC’s opposition research file on Donald Trump. The next day DCLeaks revealed the first set of hacked DNC emails, which had Cyrillic characters and Russian urls in their metadata.


Guccifer.png

First release of Hacked Russian Materials

Most of this month was filled with rumors that the DNC had spiked their support for Bernie Sanders as emails of various DNC staffer criticizing him were revealed.

On June 27, Bill Clinton had his tarmac talk with Loretta Lynch, fueling rumors that the DOJ might be improperly influenced in their investigation of Hillary Clinton’s email server. By July 1, Trump had closed the gap to 4 points (44 percent to 40 percent).


Tarmac.png

Bill Clinton talks to Loretta Lynch

Some have argued that the following actions by Comey were prompted by his concern that the DOJ had been compromised by this discussion between Bill Clinton and Lynch, even though Lynch had promised to implement whatever suggestions were brought forth by the FBI.

On July 5, James Comey announced that the investigation of Clinton’s server was complete and made a splashy public announcement declaring she was “reckless,” and then testified before Congress on the July 7, at which point Clinton led by 5 points (45.2 percent to 40.9 percent).


Emails.png

Comey exonerates Clinton, then blasts her in press conference & testimony

Although Comey had stated Clinton didn’t violate any laws, his criticism of her was quite devastating as her lead shrank to 3 points (43.1 — 40.4) by July 14.

The Republican National Convention ran from July 18-21, which naturally had a further tightening effect on the election. One day after the end of the RNC, WikiLeaks revealed their first set of DNC emails. Russia Today, Sputnik News, Breitbart, Newsmax and other right-wing outlets began doing incessant stories about every little negative revelation in them, as well as numerous false and inflammatory rumors.

Such as:

'The Obama administration was actively supporting Al Qaeda in Iraq, the terrorist group that became the Islamic State.' — Donald Trump

Hillary Clinton 'savaged' the dignity of people who accused Bill Clinton of sexual harassment or assault 'and shamed them.'— Rebuilding America Now

Attorney General Loretta Lynch 'was a member of Bill Clinton's defense team during the 1992 Whitewater corruption probe.'— Viral image

When Michelle Fields 'found out that there was a security camera, and that they had her on tape, all of a sudden that story changed.'— Donald Trump

'State Department officials actually directed contracts for the Haitian recovery after the earthquake to friends of the Clintons.' — Mike Pence


DNC.png

RNC Completes, DNC Begins — Trolls and Bots attack

As the Democratic Convention began on July 25, Trump had moved into a 1-point lead of 45.6 percent to 44.7 percent following his RNC bump. Because of the brouhaha from the WikiLeaks email dump, DNC chair Debbie Wasserman-Schultz resigned and Donna Brazille took over while RT/Sputnik heightened and amplified rumors about the DNC “rigging the primary” for Clinton and against Sanders. By the end of the convention on July 27, Trump asked Russia to hack and/or release Hillary’s missing emails. The next day Sputnik news claimed that WikiLeaks had proved the primary and DNC was rigged against Sanders.

Despite all this clamor, Clinton recovered from the bump that had been granted to Trump by the RNC, eventually regaining a 7-point lead of 47.6 percent to 40.5 percent by August 5.


Bump.png

Clinton Bump after DNC

In June just before the conventions a Russian Think Tank had released a blue print for their strategy to impact the election which involved the use of Russian State media, social media bots and paid online trolls to project confusion, turmoil and the narratives they preferred to promote.

WASHINGTON (Reuters) - A Russian government think tank controlled by Vladimir Putin developed a plan to swing the 2016 U.S. presidential election to Donald Trump and undermine voters’ faith in the American electoral system, three current and four former U.S. officials told Reuters.

They described two confidential documents from the think tank as providing the framework and rationale for what U.S. intelligence agencies have concluded was an intensive effort by Russia to interfere with the Nov. 8 election. U.S. intelligence officials acquired the documents, which were prepared by the Moscow-based Russian Institute for Strategic Studies [en.riss.ru/], after the election.

...

It recommended the Kremlin launch a propaganda campaign on social media and Russian state-backed global news outlets to encourage U.S. voters to elect a president who would take a softer line toward Russia than the administration of then-President Barack Obama, the seven officials said.

During the summer they put this plan into action.

Russian Twitter bots that year reached 1.4 Million people. And what’s even worse is that fake Russia Facebook bots reached as many as 126 Million people.

WASHINGTON — Russian agents intending to sow discord among American citizens disseminated inflammatory posts that reached 126 million users on Facebook, published more than 131,000 messages on Twitter and uploaded over 1,000 videos to Google’s YouTube service, according to copies of prepared remarks from the companies that were obtained by The New York Times.

In its prepared remarks sent to Congress, Facebook said the Internet Research Agency, a shadowy Russian company linked to the Kremlin, had posted roughly 80,000 pieces of divisive content that was shown to about 29 million people between January 2015 and August 2017. Those posts were then liked, shared and followed by others, spreading the messages to tens of millions more people. Facebook also said it had found and deleted more than 170 accounts on its photo-sharing app Instagram; those accounts had posted about 120,000 pieces of Russia-linked content.

Previously, Facebook had said it identified more than $100,000 in advertisements paid for by the Internet Research Agency.

Spending $100k to reach 129 Million people seems to me like quite a bargain. Also these ads didn’t randomly appear for everyone, they were specifically targeted by partisan region and interest.

Between June 2015 and August 2017, millions of Americans were exposed to Facebook ads and posts generated by Russian operatives who sought to influence voter behavior and exploit divisions in American society on hot-button issues. A number of the ads released during the House Intelligence Committee hearing on Nov. 1 reveal how the Russians used Facebook’s advertiser tools, as well as free posts, to target people by their interests, political leanings, location, age and other traits.

The metadata released by lawmakers show how Russians zeroed in on numerous groups — people in Texas or Ferguson, Mo., supporters of Muslims, and those who oppose Muslim immigration. Each were targeted with different messages for different reasons.

The question which remains is — did any of this move the needle, and if so — by how much and where? It would be interesting to google analytics data which ads were placed in which markets at which times in order to map and then analyze how they may have influenced voters with before and after polls, but as yet such detailed mapping and polling isn’t available.

And then there were the trolls who worked out of building in St. Petersburg which operated the bots and fake accounts which caused the stories fed by the Kremlin via RT and Sputnik directly into the bloodstream of Right-Wing media.



On the outskirts of St. Petersburg, Russia, inside an unassuming office building, people who claim to have worked there say Internet trolls are hard at work, exploiting America's deepest divisions.

...

Lyudmila Savchuk, a journalist, told Harris that she'd gone undercover to work in the troll factory for about $700 a month. She said that as part of her job, she'd invent fictional characters and then post under their names with topics carefully selected by her bosses.

'Their favorite topics were guns, immigrants and homosexuality,' Savchuk said. 'The kind of topics that could invoke blind and negative emotions.'

An independent, liberal Russian TV channel also spoke to a man named Alan Beskaev who said he'd worked in the unit that specifically targeted America.One minute, he said, 'you needed to be a redneck from Kentucky and then later you had to be some kind of white dude from Minnesota. ... And then in 15 minutes, you need to be from New York, writing something in black slang.' Beskaev said some of his colleagues even traveled to the U.S. to do research.

Another former troll said in an interview published on Oct. 16 that he worked in a so-called 'troll factory' in St. Petersburg for around 18 months until early 2015. He said that during his time there, trolls were instructed to watch the Netflix series 'House of Cards' to improve their English and strengthen their knowledge of U.S. politics.

They had even put together a phony sex-tape using a Hillary Clinton look alike and a black man — who was supposed to be Barack Obama — according to Beskeav.





Mr Baskaev, now a teacher in Thailand, appeared in an on-camera interview with independent Russian television station TV Rain. A former employee of a Russian “troll factory” has said that a Hillary Clinton lookalike and black man were hired by his company to make a sex tape during the 2016 US election.

The company, according to the first former employee to go on record Alan Baskaev, ran popular Twitter accounts that would promote then-candidate Donald Trump’s campaign officials and surrogates.

The Internet Research Agency also ran websites in favour of Mr Trump during the election, the Daily Beast reported.

As Trump continued to pound Clinton over her emails, the alleged corruption of the DNC as exposed by WikiLeaks along with the bots and trolls caused that lead to slowly shrink until it was down to just 1 point at 44.9 percent to 43.9 percent by Sept 15, just prior to the start of the debates.


Debates.png

Access Hollywood and Podesta emails break during Debates

The first debate between the candidates was on September 28, after which Clinton’s lead improved to 2.5 points at 47.5 percent to 45.0 percent by Oct. 1. Then after the second debate on Oct. 10, Clinton again stretched her lead back to 7 points at 48.1 percent to 41.4 percent, at which point things began to go sideways as just three days prior on Oct. 7 several things happened all at once.

Homeland Security reported that the Russian hacking effort had been specifically directed by the Kremlin. The Access Hollywood tape was Released. The John Podesta emails were first released by WikiLeaks ,which then proceeded to do a daily drop of them for the next month. Clinton attempted to highlight the Homeland Security report during the third debate on Oct. 21 by stating 17 intelligence agencies had confirmed Russia’s attempt to influence the election. Trump denied it, saying “nobody knows” and it could be “China, North Korea or a 300-pound fat guy” (which was also a Sputnik News talking poin). Trump also countered the growing allegations of sexual assault against him by parading all the women who had made harassment and assault allegations against Bill Clinton.

Contrary to Trump’s claims, the intelligence assessment did not prove that the Russia attacks had “no effect.” In fact, intelligence sources didn’t even attempt to answer that question.

That review did not attempt to assess what effect the Russian efforts had on the election, despite the fact that “Russian intelligence obtained and maintained access to elements of multiple US state or local electoral boards.” According to the Department of Homeland Security, the assessment reported reassuringly, “the types of systems we observed Russian actors targeting or compromising are not involved in vote tallying.”

No, those systems weren’t involved in tallying the votes, but that doesn’t mean that attacking them had no impact at all, particularly on voter registration.

After the debates Trump started saying the only way he could lose Pennsylvania was because of “voter fraud,” and mentioned the information revealed by WikiLeaks 164 times over the next month.

By the end of the month instead of keeping or extending her post-debate lead, Trump managed to close on Clinton’s 7-point lead back down to within 4.6 points at 47.1 percent to 42.5 percent.


Comey-Letter.png

Comey Letter released

At this point you had the James Comey letter to Congress on Oct 26 about Huma Abedin’s emails found on Anthony Weiner’s laptop several weeks earlier, which was leaked almost immediately, presumably by Jason Chaffetz’s office. Clinton dropped 3 to 5 percent in the polls, Trump pulled within 2 points (47.0 percent to 45.5 percent on Nov. 1) and she didn’t begin to recover until the final days before the election, at which point it was far too late.

Nate Silver argues that if not for the Comey effect Clinton would have won, but even he doesn't propose that this happened entirely on its own.

Clinton woke up on the morning of Oct. 28 as the likely — by no means certain — next president. Trump had come off a period of five weeks in which he’d had three erratic debates and numerous women accuse him of sexual assault after the “Access Hollywood” tape became public. Clinton led by approximately 6 percentage points in national polls and by 6 to 7 points in polls of Michigan, Wisconsin and Pennsylvania. Her leads in Florida and North Carolina were narrow, and she was only tied with Trump in Ohio and Iowa.polls-only forecast as of 12:01 a.m. on Oct. 28.

But it was a pretty good overall position.

Her standing was not quite as safe as it might have appeared from a surface analysis, however. For one thing, there were still lots of undecided voters, especially in the Midwest. Although Trump had a paltry 37 percent to 38 percent of the vote in polls of Michigan, for instance, Clinton had only 43 percent to 44 percent. That left the door open for Trump to leapfrog her if late developments caused undecideds to break toward him. Furthermore, in the event that the race tightened, Clinton’s vote was inefficiently distributed in the Electoral College, concentrated in coastal states rather than swing states. While she had only an 11 percent chance of losing the popular vote according to FiveThirtyEight’s forecast that morning, her chances of losing the Electoral College were a fair bit higher: 18 percent.

It may be impossible to fully isolate the specific impact of Russia’s hacking and fake news from all the other events, and it deserves somewhat more focused and detailed analysis than I’ve given it here. But the final question comes down to the three key states where Clinton had been projected to win, but did not: Michigan, Wisconsin, and Pennsylvania.

Other factors have to be mentioned, such as the fact that in Wisconsin the new voter ID bill may have disenfranchised as many as 200,000 Democratic voters in a State that Trump only won by 22,748 votes.

Prior to the 2016 election, Eddie Lee Holloway Jr., a 58-year-old African-American man, moved from Illinois to Wisconsin, which implemented a strict voter-ID law for the first time in 2016. He brought his expired Illinois photo ID, birth certificate, and Social Security card to get a photo ID for voting in Wisconsin, but the DMV in Milwaukee rejected his application because the name on his birth certificate read “Eddie Junior Holloway,” the result of a clerical error when it was issued. Holloway ended up making seven trips to different public agencies in two states and spent over $200 in an attempt to correct his birth certificate, but he was never able to obtain a voter ID in Wisconsin. Before the election, his lawyer for the ACLU told me Holloway was so disgusted he left Wisconsin for Illinois.

Holloway’s story was sadly familiar in 2016. According to federal court records, 300,000 registered voters, 9 percent of the electorate, lacked strict forms of voter ID in Wisconsin. A new study by Priorities USA, shared exclusively with The Nation, shows that strict voter-ID laws, in Wisconsin and other states, led to a significant reduction in voter turnout in 2016, with a disproportionate impact on African-American and Democratic-leaning voters. Wisconsin’s voter-ID law reduced turnout by 200,000 votes, according to the new analysis. Donald Trump won the state by only 22,748 votes.

Generally speaking, states that adopted such laws saw nearly a 2 percent drop in turnout.

Though Wisconsin saw the most dramatic reduction in turnout among voter-ID states, it was reflective of a worrisome broader national trend.

In states where the voter identification laws did not change between ’12 and ’16, turnout was up +1.3%. In states where ID laws changed to non-strict (AL, NH, RI) turnout increased less, and was only up by +0.7%. In states where ID laws changed to strict (MS, VA, WI) turnout actually decreased by – 1.7%.

Dropout rate in voter restrictive states

There’s also the point that in Detroit, Michigan, which is clearly a Democratic stronghold state, there were dozens of registration machine failures.

More than 80 voting machines in Detroit malfunctioned on Election Day, officials say, resulting in ballot discrepancies in 59% of precincts that raise questions about the reliability of future election results in a city dominated by Democratic and minority voters.

“This is not the first time,” adds Daniel Baxter, elections director for the city. “We’ve had this problem in nearly every election that we administer in the city of Detroit.” Baxter says that the machines were tested for accuracy before election day in accordance with state and federal guidelines, but that sometimes the machines “hit up against each other and malfunction” as they’re being transported to the precincts.

The machines were optical scanners, meaning they registered and counted the votes marked on paper ballots. Many of the machines jammed over the course of election day, perhaps because Michigan had a two-page ballot this year, which meant that paper ballots were collected but inconsistently recorded by the machines. Michigan does not have early voting, so any mechanical malfunction would necessarily happen on election day, since that’s the only day the machines are used. That’s why so many machines malfunctioned at the same time. “You don’t expect a laptop to last 10 years, and you shouldn’t expect a voting machines to last 10 years,” says Detroit City Clerk Janice Winfrey.

Michigan hadn’t been won by a Republican presidential candidate since 1988, but in this race Trump squeaked out a “win” by just 11,000 votes. Detroit had been having this registration machine problems for some time and there’s no current proof that Russia’s multiple attacks on voter registration systems around the nation impacted them. But then again, no one has really analyzed this issue, either.

Similarly, there isn’t much information on whether there was any direct impact from the Russian attack on voter registration in Pennsylvania (where as shown below there was a fairly low rate for counting provisional ballots) which was the third key state to deliver the election to Trump, where he “won” by just 68,000 votes.

We don't really know much about how their attacks effected voter registration, although reports for the NSA indicate they certainly tried.

The NSA has now learned, however, that Russian government hackers, part of a team with a “cyber espionage mandate specifically directed at U.S. and foreign elections,” focused on parts of the system directly connected to the voter registration process, including a private sector manufacturer of devices that maintain and verify the voter rolls. Some of the company’s devices are advertised as having wireless internet and Bluetooth connectivity, which could have provided an ideal staging point for further malicious actions.

Other reports indicate that Russian hacks which may have impacted voter registration systems in 39 states, but exactly how deeply this may have affected registrations which may have greatly slowed the ballot processing at affected precincts causing some potential voters to quit and leave early is still not clear.

Also the impact of changing the voter rolls could be devastating simply by specifically deleting Democratic voters from the list and causing those person to use provisional ballots which will only be counted if the results of incredibly close. As a matter of fact 220,000 votes like these weren’t counted during the primaries.

To determine how many people’s ballots didn’t count this year, HuffPost asked nearly every state that held a presidential primary this cycle how many provisional ballots were cast and how many of those provisional ballots were ultimately counted. Some states partially count ballots cast outside of the voter’s precinct by counting selections for state and federal offices, but not local races.

Here is a graphic showing what we found; it doesn’t include data from states where only a minuscule number of provisional ballots were recorded, like Vermont and Kentucky.

HuffPost got statewide data from 23 of the states that held primaries this year, plus data from Arizona’s largest county, Maricopa; data from Kansas, whose GOP caucus weirdly issues provisional ballots — caucuses don’t typically involve ballots; and data from New York City, where the chaotic Democratic presidential primary resulted in an unusually high number of provisional ballots submitted and rejected. (Public affairs officials with the New York State Board of Elections ignored repeated emails and calls from HuffPost.)

Just this week Democratic members of House Oversight wrote a letter to Chairman Gowdy requesting that he subpoena Homeland Security for documents related to Russian attacks on our elections systems in dozens of states and he hasn’t bothered to even answer yet.

is it possible that a combination of effects—Trump’s appeal to the Rust Belt and nativism, Clinton’s email troubles and lack of clear economic message, voter suppression and registration problems, together with efforts by Russia to push the election in favor of Trump—all combined to create the results we all saw on November 8?

Well, of course it could. If you shoot a guy, stab him, then throw him off of a roof, is it the fall that killed him or everything combined? Could he have survived the impact on the ground if he hadn’t already been wounded?

Nate Silver’s analysis contends that Comey letter alone in the best case scenario could have still changed the outcome of the election if it only made a difference of just 1 percent, but then that would have only been possible because the race was already so close.

Nonetheless, Clinton lost Michigan, Pennsylvania and Wisconsin by less than 1 percentage point, and those states were enough to cost her the election. She lost Florida by just slightly more than 1 point. If the Comey letter had a net impact of only a point or so, we’d have been in recount territory in several of these states — but Clinton would probably have come out ahead. I call this the “Little Comey” case — sure, the Comey letter mattered, but only because the election was so close.

ADJUSTED VOTE MARGIN
CLINTON VOTE MARGIN SMALL COMEY EFFECT* BIG COMEY EFFECT* CLINTON’S ELECTORAL VOTES HAD SHE WON
Michigan -0.2 +0.8 +3.8 248
Pennsylvania -0.7 +0.3 +3.3 268
Wisconsin -0.8 +0.2 +3.2 278
Florida -1.2 -0.2 +2.8 307
Nebraska’s 2nd C.D. -2.1 -1.1 +1.9 308
Arizona -3.5 -2.5 +0.5 319
North Carolina -3.7 -2.7 +0.3 334
Georgia -5.1 -4.1 -1.1 350
Ohio -8.1 -7.1 -4.1 368
Texas -9.0 -8.0 -5.0 406
Iowa -9.4 -8.4 -5.4 412

Even a small Comey effect could have cost Clinton the 270 electoral votes she needed to win

*Adjusting for a small Comey effect adds 1 percentage point to Clinton’s vote margin. A big effect adds 4. Hypothetical scenario starts with the 232 electoral votes Clinton actually won, ignoring faithless electors.

The question we really need to ask is: Without the impact of all the negative news against Clinton that was provided to Trump by Russia and WikiLeaks (which helped bring her 7-point post-convention lead down to just 1 point and her post-debate lead down to just 4 points), could she have survived all the other elements (including the release of the Comey letter in October) with what would have still been a 4-point margin instead of just 1.5 points on November 1? Would she have been able to retain her projected leads in Michigan, Wisconsin, and Pennsylvania despite various voter suppression efforts and registration problems?

No one can say they know for sure, but we all deserve to know and someone should be focused on this issue.

Russia ran a similar attack strategy against the Brexit Vote, the 2017 German election, the French Election infrastructure and the Catalonia Separatist vote in Spain, while CIA Director Mike Pompeo says he expects them to try again during our 2018 midterms.

LONDON (Reuters) - CIA Director Mike Pompeo said Russia will target U.S. mid-term elections later this year as part of the Kremlin’s attempt to influence domestic politics across the West, and warned the world had to do more to push back against Chinese meddling.

...

In an interview with the BBC aired on Tuesday, U.S. intelligence chief Pompeo said Russia had a long history of information campaigns and said its threat would not go away.

All indications are this just might happen against in 2018. We need to be ready.