Government Spending

Nate Silver: Stingy Lunch Date?


Nate Silver, every girl's ideal lunch date.

You may have enjoyed the recent tango between Nate Silver and Reason Contributing Editor Veronique de Rugy over how much Democratic party affiliation factored into stimulus awards under the American Recovery and Reinvestment Act.

De Rugy's original work, drawing on the woefully incomplete data set available at the Treasury Department website, suggested that districts represented by Democrats received vastly higher stimulus awards than those represented by Republicans.

Veronique de Rugy is a known French person.

You can see de Rugy testifying about the party-specific pattern of stimulus awards before the House Committee on Transportation and Infrastructure. [Spoiler: Rep. James Oberstar (D-Minnesota) steals the show by speaking French with what is certainly one of the most mignon accents ever to come out of the Gopher State.]

Silver countered that many state capitals—where federal stimulus funds would logically stop on their way to whatever local factory makes big green signs that read "Project Funded By the American Recovery and Reinvestment Act"—are in Democratic districts. Thus these districts take up such a large portion of the overall stimulus awards that the study should have controlled for them. Silver's polite post on the matter concluded with a challenge:

If de Rugy follows my recommendations—excludes state capitals, accounts for a broader array of demographic variables, and evaluates unemployment rates at the district level—and still finds a statistically significant positive relationship between the distribution of stimulus funds and whether the district elected a Democrat to Congress, I will buy her and three of her colleagues lunch anywhere in Washington or New York City.

The sign alone created dozens of jobs.

The multi-talented de Rugy responded with a post calling Silver a "respected blogger." Three-time Oscar® nominee Silver followed up with a post calling de Rugy "smart and capable." The chirping classes were less civil, but the disagreement between these two mathletes gets to some important points, including the disappointing shallowness of data from Treasury.

Now de Rugy has published a new version of the study, making the adjustments Silver suggested. Her findings:

On average, Democratic districts received 1.53 times the amount of awards that Republicans were granted. The average number of awards per Republican district is 112, while the average number of awards per Democratic district is 171…

Democratic districts also received 2.65 times the amount of stimulus dollars that Republican districts received ($122 billion vs. $46 billion). Republican districts also received smaller awards on average… In total, Democratic districts received 73 percent of the total stimulus funds awarded and Republican districts received 27 percent of the total amount awarded…

Regression analysis (ordinary least squares) was used to explore the predictive power of the various indicators described in the section above. The political variables considered here included the district representative's political party, tenure in office, leadership position, membership on the appropriations committee, and voting in the most recent presidential elections, as well as the inclusion of the state's capital within that district. The analysis finds that a district's representation by a Republican decreases the stimulus funds awarded to it by 27.9 percent. This result underscores the findings from the previous Stimulus Facts report.

This effect is statistically significant at the p < .015 level (see table 1). The regression analysis does not seek to explain (nor does it explain) precisely how funds were allocated (adjusted R2 = .400). That would require a more complete dataset than has been used for these results or is available through"

Silver had not responded—an understandable move when faced with the prospect of eating lunch in the presence of four libertarians. But his critique underscores the importance of getting more complete data on where each dollar of stimulus originated and where it finally ended up. And also how much of the dollar was left when it reached its destination. Noting that state capital districts go on to disburse some portion of the funds they receive is important, but not dispositive. It's a law of economics, or should be, that a suitcase full of money gets emptier every time it has to make a stop.

NEXT: What Americans Think About the Budget

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. I note some weasel words in Silver’s challenge: “…a statistically significant positive relationship…”

    1. That’s not all that weaselly, statistically significant isn’t too hard to define. The important P values are low.

  2. It’s always nice to win a battle of the models. I enjoyed your working paper.

    Thanks and enjoy lunch.

  3. I never agreed to 3 colleagues on top of the vig! Am I something special? Some sort of schmuck on wheels?

  4. Silver had not responded — an understandable move when faced with the prospect of eating lunch in the presence of four libertarians.

    I think this is the most extraordinary collection of talent and of human knowledge that has ever been gathered together at lunch anywhere in Washington or New York City ? with the possible exception of when Thomas Jefferson dined alone.

    1. with the possible exception of when Thomas Jefferson dined alone.


  5. > I will buy her and three of her colleagues lunch anywhere in Washington or New York City.

    Umm, the Rainbow Room (NYC) is especially pleasant in the springtime. Ask for a table in the pool room–and no, the room doesn’t feature a billiards table.

    1. and no, the room doesn’t feature a billiards table.


    2. Didn’t they close the Rainbow Room a couple years ago? Am I just making that up?

  6. Why…am I paying for signs, the sole purpose of which, is to inform me that I’m paying for signs?

    1. How else would you know you’re paying for signs?

      1. What is the penalty for defacing one of those signs with a carton of eggs? Not that I would have that opportunity, living in a Republican state.

        1. Since eggs wash off pretty easy (compared to spray paint at least), does that really count as defacing?

          1. A former girlfriend’s car got egged and it baked in the sun before she had time to wash it off. Never did come out of the paint. Something to do with proteins, I think. Or maybe justice.

            1. No Penalty at all. By egging the signs you are creating jobs for people who have to clean or replace the signs. You will be a hero and get an award from the President.

        2. Sure, so long as the eggs on my sign were subsidized using my money. Wouldn’t want to break the wealth-destruction circle…

    2. sign manufacturer lobby?

    3. Next time you see some bad potholes, go back and get some cardboard, a stick, some tape, and a magic marker and post a sign reading “These potholes courtesy of TARP and ObamaCare.”

  7. I’d do an additional study … stimulus dollars relative to the margin of victory for either party.

    I.e. Do “swing” districts get disproportionate amounts of funding in the hopes of swaying voters. Or do solidly “Red” or “Blue” districts get less since there’s no chance of changing the election results?

    1. That would be a good variable. Something to account for how contested a district is and the relation to stimulus money.

      That could be a pretty telling variable.

      1. Yeah, swing states for presidential elections too.

        There could be some relationship between Obama’s margin of victory and the stimulus money (per capita) directed to the state. In an effort to secure the state as “Blue” in future presidential elections.

    2. Do you really think they’re smart/disciplined enough to give money where they need it as opposed to where the crusty oldtimers represent?

      1. You think those are separate goals? (need being defined as: doing what gets us more votes)

  8. I am offering, here and now, to take Veronique to lunch, and, if things turn out well, to breakfast much later.

  9. I’m guessing by those two pics that she could clean his clock.

    1. The pics with the original post were even more striking – Silver looked like he’d just evolved out of the primordial ooze, or perhaps was in the midst of a three-week coke and booze orgy in Tijuana.

  10. Was all the data up and GMU when Silver made his claims? If so it was a pretty stupid thing to do without running a few regressions of your own.

  11. What’s up with David Weigel’s negative portrayal of de Rugy’s study? Aren’t they colleagues or something?

    “Massively flawed study” of my ass

    1. Where did he say that?

      1. Never mind, I totally missed it.

    2. Didn’t you know the econometric and stats requirements in J-School are fucking horrendous? Not to mention the rigorous modeling and analytical problem solving requirements of journalism. I mean hell I’m amazed he didn’t set up his own models and regressions to counter this. He could probably do it in his head and just shit out the data.

    3. There was a flaw that undercut the premise of the study and dramatically changed the result. What would you call it?

      1. Now that we know you read the comments, we’re going to be even meaner.

      2. I would call it a working paper. Just what the author called it. I would also not represent the change in the model as moving the important coefficients from statistically significant to being insignificant. Just because the coefficient decreased neither decreases statistical significance nor does it violate the initial hypothesis. The shift was recognized and stated.

        Your snide little quip was pure hackery. You got called on it and proven wrong, or at the least proved you understand statistics and econometrics about as much as you do fashion and how to use a gym membership. Now it’s time to man up and apologize or start churning out your own papers and models. Or just STFU and let the smarter people play while sticking to reporting what senator is banging what intern.

  12. Dave Weigel really is an insufferable little shit, isn’t he?

    1. I’m disappointed. One thing I liked about DW in the brief time he was with us is he seemed to know when to shut the Hell up and offer up some prog rock more times than not. Now he has gone to that big bustling S&M club of beltway journalism where the young j-school cubs are made to lay naked, tied, and stretched out on a bed of nails while a sweaty David Broder rolls his flap jacky man titties in their faces he has forgotten all of those lessons we taught him concerning basic human dignity that made him almost tolerable.

      1. “…flap jacky man titties…”

        That’s some good shit!

      2. The only good thing Weigel ever did at Reason was put his name on his articles as a sign that the following paragraphs were going to be substance-free.

    2. Have you seen his “blogroll”?

      Marc Ambinder
      John Cole
      Ana Marie Cox
      Kevin Drum
      Economy Watch
      The Fix
      The Hearing
      Hotline On Call
      The Independent Streak
      Ezra Klein
      Lawyers, Guns & Money
      Julian Sanchez
      Dan Savage
      Ben Smith
      White House Watch
      Matthew Yglesias

      No shoutout to his former Kochtopus masters.

      1. Yeah, that’s some bunch of “libertarians” he’s got there. Are there any on that list other than Julian Sanchez? Why was Weigel ever at Reason?

        1. Julian Sanchez isn’t much of a libertarian.

      2. Have you seen his “blogroll”?

        I totally misinterpreted that after reading the flap jacky man titties comment.

    3. Speaking of Weigel, I saw him on TV the other night, and he’s still wearing clothes that are two sizes two small. What’s up with that?

      1. That should be “too small.”

      2. I bet he shops at American Apparel. What a fucking degenerate.

  13. There’s on of those big, fucking, green signs in my town. I’m going to plaster it with Ron Paul stickers, take a picture of it, and post it in the comments section of the local newspaper everyday until I get blackballed, or arrested.

    1. Those signs are retarded. Obvious effort to buy political support.

      1. I had never seen one in person until the other day. They are fucking huge. About 3 to 4 times larger than the caution signs alerting drivers to slow down for the construction zone. I guess the effort to gain political support is much more important than the safety of the workers.

        1. I know.

          I wish someone would start a campaign to alter the signs to read “This SIGN paid for by the American Recovery and Reinvestment Act”

          1. How about “This Sign paid for by you, the American taxpayer / sucker”?

            1. Eggs are cheaper.

            2. “This Sign paid for by you, the American taxpayer. Sucker.”


            3. nah. my version is subtler and funnier. I bet in some places it wouldn’t be noticed for weeks, and thus last longer as an inside joke.

    2. Have you seen the bullshit census ad with the guy running around saying ‘the census is great so we can give lots of free money to people’?

      Fucking propaganda paid for with tax dollars.


        First off, i can’t believe it will really get more responses. At least not enough to justify the cost.

        Secondly, i have a beef with a specific part of their propaganda. “We need the census to know how many schools to build!” Yeah, because it would be totally impossible to use current enrollment records to figure out how many new schools we need. No no no no no, we NEED a whole ‘nother information collection to figure that out!

        While I’m on it, how about ditching commercials for the armed forces? Is there anyone in America who saw one of those and was like “Shit, I like my Job but that looks WAY fucking better!!!!” Nah, people join the armed forces because

        1) it runs in the family
        2) they feel a need to do it (national pride, being a “protector”, or just being plain homicidal)
        3) They have no other options

        People in those 3 catagories KNOW the Armed Forces are there, and I’m sure they can find the phone number in their local yellow pages.

        But hey, lets keep buying prime time advertising on popular cable channels. It’s not a waste of money, not at all…

        1. Nor is it even a waste of *Billions* of units of money! Not at all…

    3. I’m not sure that these signs are having the positive effect that they want.

      For instance, there is a very important stretch of road used by tons of MN-ans getting out of the Twin Cities each weekend to go “Up North” for the weekend. It was in so-so shape. Now it has been ripped up for nearly a year causing huge delays in everyone’s vacation plans.

      Aren’t all those people who have been inconvenienced for a marginal improvement in a road more likely to vote against the idiots who pulled this off?

      1. Roads. One of the original/true purposes of gov’t. And they can’t even do that with any measure of effectiveness.

    4. We rednecks prefer bird shot and buck shot. This might be one of the reasons there aren’t too many of these signs in the middle of nowhere.

      1. Although I suppose one could pattern a shotgun with a government road sign they are commonly used to sight in a rifle. A properly shot-up road sign should show a few “flyers” and a 2 or 3 shot “group”, often on/in an “O” or “0”. Sometimes you’ll see a vertical string with 2 shots “on target”.

        1. I’ve met by tag allowance in street signs. I was going for the drunken sign destruction method. Which if using a rifle require tannerite

  14. Nate loves Veronique! OMG!

  15. Controlling for state capitols? They tend to be magnets for Democrats. You might as well control for cities near military bases for the Republican side of that ledger sheet too.

    1. Also need to exclude anyone within 2000feet of a school, church, or daycare center.

      There, now that we’re counting just a few cityblocks worth of America, should come up with the numbers Silver wants.

  16. Silver had not responded — an understandable move when faced with the prospect of eating lunch in the presence of four libertarians.

    I think the more fearful prospect is paying for it.

  17. There are stimulus signs around SoCal where you’d be pressed to see actual construction. I think that’s got people thinking how much of a waste of money it is.

  18. Wow, Jim Oberstar’s French is REALLY good. Like, accentless (or close to it?). Looking at Wikipedia I see that he was a French teacher for the Marines for four years, but that was decades ago, and he still has a better accent than any of my ~10 non-native French teachers.

    1. I picked up french really well in High School. The french people i talked to when i went to France thought i was a native speaker. To the point where i had to make them slow down their responses: i could speak it well, but understanding it was a different animal.

      My Point: Ten years later I can still sound just like a frenchman, except i forget most of the verbs and nouns, so all i can talk about is food or the bathroom.

      So, grats to Oberstar for not forgetting what he knew.

      1. Also: really wish now that i’d taken spanish instead….

  19. Silver had not responded — an understandable move when faced with the prospect of eating lunch in the presence of four libertarians.

    Hm. Probably he doesn’t care all that much for breast of crow, seasoned with bitter herbs and washed down with turned white whine.

    1. Or what a Scotsman would call a mildly pleasant lunch.

  20. My objections to Baseball Prospectus used to be long and tedious. Now I have a simple explanation:

    “They don’t believe that one party who controls all three houses and votes to spend $1 trillion will spend more in their districts without “peer-reviewed” research . . . even when the data to do so is for no apparent reason unavailable and the data available points overwhelmingly in that direction.”

  21. I can’t wait for the 98% tax rates on Democrats once the Republicans are in power.

    Dems say they like taxes? Well then they shouldn’t have any problems when they are rammed down their throats.

    The teabaggees will say: ‘hmmm mmm mm mmmmmm mmmmmmm’ which translates as ‘I want more!’

  22. What’s with Dave Weigel being a toolbag and calling de Rugy’s study “massively flawed”? Now I understand all the Weigel-bashing here. Why is this Obama-voting moron (redundant, I know) writing about the “inside scoop” of conservative news? I’m almost tempted to mock the fat turd for his portliness too, but tend to be passive-aggressiveness with such insults.

    1. I’m almost tempted to mock the fat turd for his portliness too…

      Now that was offside, wasn’t it? He’ll be crying himself to sleep tonight on his huge pillow.

    2. He’s going hipster. Maybe he got himself a hipster girl

    3. Weigel is on the “anti-government Hate-Watch beat” for the WaPo.He probably regales his colleagues with war stories from his undercover stint in the lion’s den hanging out with dangerous right-wing extremist radicals like Katherine Mangu-Ward and Brian Doherty.

      (I do give him props for top-notch campaign reporting when he wasn’t totally shilling for the DemocRAT party)

  23. Shouldn’t there be an urban-rural test as well? Urban areas have higher infrastructure needs and would presumably pull in more infrastructure spend. They also tend to vote democratic as well. It would be interesting to see if Orange County got fewer stimulus dollar per capita than Westchester County, for example.

    1. “Flaming Mo” would be an awesome alter troll ego.

  24. Agree with the main column. But a bit of minor fact-checking about the reference to Nate Silver being a three-time Oscar nominee. Really? Could you have him confused with the late great (and politically courageous) actor Ron Silver? Nate Silver is mentioned in Google for his columns and blogs predicting the Oscar races in recent years — but not actually being nominated for any. He’d have to be an actor or film-maker to be nominated, I suspect.

  25. Dustin Hoffman was in Star Wars.

Please to post comments

Comments are closed.