My Hopefully Final Word (Well, More Like a Whole Lotta Words) on the Krugman Election Controversy, Responding to the Half-Hearted Defense of Krugman Proposed by Tom Maguire
Tom Maguire thinks he has come up with an explanation for Paul Krugman’s recent assertion that two of three recounts of undervotes would have given the 2000 election to Al Gore. If Maguire is right, then Krugman’s assertion is still dishonest — but in a slippery and underhanded fashion, rather than a blatant and easily disproved fashion.
That sounds more like the Krugman we know and love! Which makes me think Maguire is on to something.
Warning: the following post contains some painful resurrected details of the 2000 post-election brouhaha. If reading those details stressed you out then, re-living them today may have similar effects. Pregnant mothers and those with weak hearts are warned that they proceed at their own risk.
The entire affair started with this August 19 Krugman column, in which Krugman falsely stated:
Two different news media consortiums reviewed Florida’s ballots; both found that a full manual recount would have given the election to Mr. Gore.
That was just blatantly false, as even Krugman admitted in a correction in his August 26 column. But that correction also raised a few eyebrows, as it appeared to once again deceptively state the results of the Miami Herald consortium’s survey of undervotes:
[T]he public editor says, rightly, that I should acknowledge initially misstating the results of the 2000 Florida election study by a media consortium led by The Miami Herald. Unlike a more definitive study by a larger consortium that included The New York Times, an analysis that showed Al Gore winning all statewide manual recounts, the earlier study showed him winning two out of three.
An August 22 column with an appended correction made it clear that the “earlier study” Krugman was referring to was an April 2001 study by a “media consortium led by The Miami Herald.” That column also made the curious “two out of three” assertion:
About the evidence regarding a manual recount: in April 2001 a media consortium led by The Miami Herald assessed how various recounts of “undervotes,” which did not register at all, would have affected the outcome. Two out of three hypothetical statewide counts would have given the election to Mr. Gore. The third involved a standard that would have discarded some ballots on which the intended vote was clear. Since Florida law seemed to require counting such ballots, this standard almost certainly wouldn’t have been used in a statewide recount.
Regular readers will remember that I was agog at reading Krugman’s pronouncement regarding the results of the Miami Herald consortium’s study. A May 2001 description of the study’s results in USA Today clearly stated that Bush would have won three of four recounts:
The newspapers then applied the accounting firm’s findings to four standards used in Florida and elsewhere to determine when an undervote ballot becomes a legal vote. By three of the standards, Bush holds the lead. The fourth standard gives Gore a razor-thin win.
The paper went on to note that the razor-thin win (3 votes) was too close to be considered reliable.
At the time, I said:
I’m just utterly flummoxed. Is a New York Times columnist just repeatedly lying to his readers about an easily checkable fact, even after getting called on it by his public editor? Or is this guy living in a parallel universe where what he is saying is true? Please, someone help me. What is going on here?
After all, Krugman’s dishonesty is usually the slippery and underhanded kind — not as seemingly blatant as this. I genuinely couldn’t figure out what was going on. I observed, not completely facetiously: “I’m really starting to wonder whether Paul Krugman is looking at a different 2001 study of undervotes by a consortium including the Miami Herald than I am.”
I e-mailed the indefatigable Tom Maguire for help. He calls ‘em as he sees ‘em and does excellent research. If anyone could get to the bottom of this, I thought, perhaps he can.
Tom appears to have found something that, if you twist it this way and that, and ignore several inconvenient facts, offers some deceptive textual support for the Krugman claim. Tom’s explanation renders Krugman’s assertion merely sleazy and deceptive, rather than blatantly false. Bingo! That’s our boy’s M.O.! Tom’s explanation has got to be right.
The explanation relies on this language from an April 4, 2001 Miami Herald article, reproduced by Maguire here. The Herald observes in the article that numerous undervotes never made it to the canvassing boards to begin with, and says:
Had all canvassing boards in all counties examined all undervotes, thousands of votes would have been salvaged in Broward County, Palm Beach County and elsewhere long before the election dispute landed in court – and the outcome might have been different, The Herald found.
In that scenario, under the most inclusive standard, Gore might have won Florida’s election – and the White House – by 393 votes, The Herald found. If dimples were counted as votes only when other races were dimpled, Gore would have won by 299 votes.
But if ballots were counted as votes only when a chad was detached by at least two corners (the standard most commonly used nationally), Bush would have won by 352 votes.
Does this explain Krugman’s “two out of three” claim that had me so mystified? Perhaps. Does it make that an honest claim? Not a chance, for many different reasons.
Most obviously, this is not the way the newspapers reported their findings — it’s an afterthought buried within the articles, based on an unrealistic hypothetical (that the courts would toss out already-completed hand recounts). But Krugman treats it as the study’s major finding. That is deceptive right off the bat.
Second, the claim overstates the certainty of what would have happened if all undervotes were counted — because the newspapers didn’t necessarily examine all the undervotes. The very same article states:
Of course, regardless of the undervote reviews, only one thing is truly clear: Precise numbers released on Election Night mask a world of imprecision and chaos.
Responding to Herald requests for undervotes, only eight of Florida’s 67 counties were able to produce for inspection the exact number they reported on Election Night. Elsewhere, there is no way to know whether the Election Night figure, the number of ballots actually inspected, or some other number is correct.
Third, Krugman’s formulation is a play on words that describes two standards favoring Bush (“two-corner detachments” and “clear punches”) as only one (“at least two corners detached”), making it seem like Bush won under only one standard. But you could just as easily describe the two Gore standards (“any dimples” and “dimples with other dimples”) as a single standard (“dimples”). With that verbal sleight of hand, you could say Bush won two of three standards (“two-corner detachments” and “clear punches”) vs. Gore’s one (“dimples”). It’s the same thing, but — presto-change-o! — Bush, not Gore, wins two of three.
What fun you can have with deceptive wording!
Of course, Krugman takes his description from the description in the article itself — but the authors of the article would probably have taken more care to separate out the different standards if they were reporting this as the study’s major finding — which they were not, though Krugman pretends they were (assuming Maguire is right to assume this article is the basis of Krugman’s assertion).
But Krugman’s most obnoxious claim is the one that gets the blood boiling the most when you finally understand the chutzpah of his assertion. That is his claim that the standard under which Bush won an undervote recount is an unreliable standard that “almost certainly wouldn’t have been used in a statewide recount.” This is the very same standard described by the Miami Herald as “the standard most commonly used nationally.” Interesting. Krugman doesn’t mention that.
The obvious question is: why would the most common nationwide standard be one that “almost certainly” wouldn’t have been used in a statewide recount? Brace yourself. According to Krugman, it’s because it that standard, the two-corner standard, “would have discarded some ballots on which the intended vote was clear.”
Clear?? These are dimples, folks. Dimples!
Dimples aren’t clear.
First, let’s remember why those undervotes never made it to the canvassing boards to begin with. From an April 5, 2001 Miami Herald article, reproduced by Maguire here:
Consistency was hard to come by. Piles of dimpled ballots never made it to the canvassing boards for scrutiny because elections workers and observers agreed among themselves that they contained no valid votes.
So this treasure trove of votes — including a significant number where the voters’ intent was supposedly “clear” according to Krugman — consisted of ballots that, elections workers and observers from both political parties agreed, showed nothing.
At the risk of giving you flashbacks that may send you into paroxyms of rage once again, let’s take a very close look at just how “clear” dimples are. All we need to do is consult the very same April 5, 2001 Miami Herald story we just quoted. This is where it gets ugly, and the bad, bad memories start to resurface. Remember, as you read this, that Krugman is saying dimpled ballots can be “clear”:
[A]s the recount battle went to court again and again, and the canvassing board members saw dimpled ballot after dimpled ballot, the basis for judging a vote evolved.
In both counties, board members started looking at the whole ballot rather than just the presidential chad in an effort to determine voter intent. In Palm Beach, the canvassing board counted dimples as votes if the rest of the ballot bore similar marks instead of clean punches.
“Generally there had to be some pattern that this was how the person voted,” said Judge Charles Burton, the chairman of the Palm Beach board. “Out of 22 votes if you just had two little dings, we wouldn’t necessarily count that.”
Broward canvassing board members Robert W. Lee and Gunzburger tended to view a dimple as a vote if there were other marks on the ballot for candidates of the same party. Lee, a Democrat and county court judge, even made a list showing which punch-card numbers corresponded to Democrats and which ones corresponded to Republicans. A quick glance at the list and the ballot would show whether the voter appeared to choose a straight ticket.
“There had to be a pattern of two or three dimples in the Democratic field for me to feel comfortable to count a dimple for Gore,” Lee said.
. . . .
Even canvassing board members acknowledge they could not be 100 percent consistent over the long days. “I’m sure there’s a few [ballots] in there now that if I went back and looked, I’d say these are votes, and if I went through the votes, I’d say some are not votes,” Burton said.
Take a deep breath, folks — we’re still not done:
The order in which ballots came before the canvassing board was another variable. If the board saw a dimpled ballot and called it for Gore, they might call the next dimpled ballot for Bush. But if a similar ballot came three hours later, it might be discarded.
“At 10 a.m. a person might be a little more conservative, and by 10 p.m they may be a little more liberal,” said LeMieux of the Broward GOP.
Oh my God. Are you remembering the horror of watching this all unfold? It’s all coming back to me, and it’s not pleasant.
So why am I putting you through this pain? Yes, I’m being cruel — but there’s a reason: we all need a reminder of just how “clear” voter intent is when punch-cards are merely dimpled — and why the Supreme Court was dead right to find an equal protection violation due to the standardless recounts at issue in Bush v. Gore. When the people responsible for counting the votes “tended to” view dimples as votes depending upon their subjective analysis of other votes on the card; when they say things like “we wouldn’t necessarily count that”; when they say that a ballot they would have counted as a vote on one day, they wouldn’t have counted as a vote the next; when the standards evolve while the count is going on — well, then, Mr. Krugman, that is not what I call “clear.”
True, you could have had a recount that simply chose a standard (for example, any mark on a chad) and counted ballots that way. But doing that doesn’t mean that the voter’s intent on a dimpled ballot would be “clear.” And, even applying such a seemingly unmalleable standard wouldn’t necessarily get you consistent results. As the same story acknowledges:
Adding pressure on boards was the inherent difficulty of assessing punch cards. A chad is no bigger than a freckle. A dimple to one person can be a shadow to another.
Tom Maguire’s valiant efforts notwithstanding, I have yet to see a scenario under which Krugman can be fairly said to have represented truthfully the results of the 2001 undervote recount. If Maguire has truly found the basis of Paul Krugman’s statement, it merely reveals another scenario under which Krugman is being dishonest. Indeed, this scenario involves multiple new layers of Krugman dishonesty. You can peel that onion all night long and never finish the job.
Of course, all it would take would be one e-mail from the public editor or Krugman himself, and we could end this game of speculation over where Paul Krugman gets his facts. Whaddya say, guys? Can you solve the mystery for us?