Keep Going…

Last month, the governor of Ohio, Ted Strickland, signed an executive order forbidding the purchase of services from overseas companies.  His reasoning is that it’s better for “recovery” that state agencies be forbidden from purchasing services from companies located overseas.  Unfortunately, Strickland is not going far enough.  Given how Ohio’s unemployment rate has been above the national average for the last several years, clearly the executive order should limit state agencies to purchasing services from companies based in Ohio.  Actually, even better, to help out poor Meigs county residents, the executive order should mandate that services be provided solely from providers in Meigs county.  In fact, Chester Township, in Meigs county, has an unemployment rate north of 14%, so the executive order should focus on there.  But we’re going to focus jobs on Township Highway 1024 instead of those State Route 248 snobs, so we’ll go there.  Heck, while we’re at it, only companies located at the corner of Township Highway 1024 and Scout Camp Road should be eligible for services.  We are going to stimulate that street corner like never before!

All joking at the expense of our soon-to-be-former governor aside, what’s interesting is the way in which people in state agencies are trying to get around this executive order, and how loosely it’s being interpreted.  Suppose you were looking at a security tool.  Imperva has one of the better tools on the market.  They have a headquarters in Redwood Shores, California and a headquarters in Tel Aviv, Israel.  Would it violate the executive order to purchase their product?  “Well, no…” say the legal people:  the executive order just says “services,” not “products.”  But what if they have help desks located in California, Israel, and the Philippines (to provide 24-hour support)?  That is an open question, even though the text of the document reads that “Despite these requirements, federal stimulus funds were recently used to purchase services from a domestic company which ultimately provided some of those services offshore.  This incident was unacceptable […]”  It sounds like this product should be forbidden, and yet there are two state agencies which recently purchased it (though probably before the stupid executive order was put into place).

What about products we currently have?  One of the best database software companies is based out of the UK, so clearly they’re a bunch of filthy foreigners who shouldn’t get our money.  Boo to them!  But there’s a legitimate chance that they can win an open competition for software procurement:  they provide good products at a very low price.  We don’t use them in our agency (I ended up selecting a different company), but our governor has made it so that they’re just right out.

One product we do own is based in Australia and—heaven forfend!—their tech support is also out of Australia.  Aside from sucking our money into their evil, down under paws, they provide a valuable service which makes our agency run more efficiently.  We use Microsoft SQL Server, which includes a large number of foreign developers, support staff, MVPs, and so on.  Apparently, I should stop using that.

Honestly, if Strickland wanted to increase the number of jobs in Ohio and stimulate the state economy, he could start by making it a better place to do business.  Ohio is currently a below-average state in terms of doing business, and is miserable at business-friendliness and its workforceOther charts have Ohio placing even worse (though apparently small businesses are better off).  Getting rid of onerous legislation and regulation, as well as lowering taxes (and spending, to balance the budget) would do a lot more to help than a half-cocked import substitution scheme.


In The Papers: The Firm That Slacks Together Works Together

Benjamin Waber, Daniel Olguin Olguin, Taemie Kim, and Alex Pentland have a very interesting piece of work, entitled Productivity Through Coffee Breaks:  Changing Social Networks by Changing Break Structure.


In this paper we present a two-phase study undertaken to experimentally study in a real world setting the effects of social group strength and how to increase the strength of groups in the workplace. In the first phase of our study we measured interactions between workers at the call center of a large bank based in the United States using Sociometric Badges. We confirmed our hypothesis that the strength of an individual’s social group was positively related to productivity (average call handle time) for the employees that we studied. In the second phase of our study we show that by giving employees breaks at the same time we increased the strength of an individual’s social groups, demonstrating that low-cost management decisions can be used to act on these results.

The paper’s goal is to measure social interactions between employees at a large call center (3).  In this case study, the authors note that this is a very large call center (10,000 employees), and despite this, teams still stagger their break schedules in order to ensure coverage.  The authors then were given permission to change around the break schedule in order to measure productivity changes.  They note that there is a positive relationship between social group strength and productivity (3), but more interestingly, that there is a causal relationship in at least one direction:  increased group strength increased productivity (4).

They were able to study this using a tool they created called a Sociometric badge.  It recognizes movements, captures nonlinguistic signals (though it does not analyze words; they prevented that functionality for privacy reasons), locates people within an area of approximately 1.5 meters, and can even detect face-to-face conversations (4).  By tracking the various teams under a single manager and operating in the same guidelines, they found that teams with a common break period performed better than staggered-period teams, because teams with common break periods allowed friends to spend more time together.

Also interesting in this paper:  the word “kith” (11, fn)  Kith is a cohesive group with common beliefs and customs.  Use that word as often as you can.

In The Papers: Reforms In Eastern Europe

I am not the biggest fan of Jeffrey Sachs, but I sat down to read Reforms in Eastern Europe and the Former Soviet Union in Light of the East Asian Experiences because I figured it would be interesting to read the words of one of the men who was very deep in the issue.  This paper is from 1995, but it holds up well.


During the past five years, there has been an important debate over the differing styles of market reforms in the formerly planned economies in East Asia versus Eastern Europe and the former Soviet Union (EEFSU). This paper puts forward three related propositions. First, the rapid growth of East Asia, compared with economic contraction in EEFSU, reflects differences in economic structure and initial conditions, rather than differences in economic policymaking. Second, East Asian gradualism could not, and did not, work in EEFSU. Third, EEFSU continues to face serious problems with an overextended welfare state inherited from the socialist period.

Sachs dismisses three theories for why East Asia (China, South Korea, Taiwan, etc.  I don’t believe that Sachs discussed Singapore here, and the example of Hong Kong goes entirely against most arguments for what the Asian Tigers did) had grown faster during the early 1990s than eastern Europe:  first, that gradual reform is superior to shock therapy; second, that economic reforms should be undertaken before political reforms; and third, that most Asian countries followed Japanese industrial policy, which was a better choice than the laissez-faire policy of the Washington Consensus (3).  All three, Sachs argues, are wrong.

It is important to note that eastern Europe and Russia started out as industrialized nations, whereas most of Asia started out as agrarian.  This left less room for capital to move to good uses in Europe (3).  Because of Soviet over-industrialization and the emphasis on terribly inefficient processes, Sachs implies that eastern Europe was actually in worse shape to begin with than eastern Asia.  In addition, there were significantly smaller welfare states in eastern Asian countries, whereas the EEFSU (Eastern Europe and Former Soviet Union) was burdened with a decades-long bias toward government handouts (4).  As a result of this, any move toward a freer economy would harm some social groups.  This is inevitable when State-owned enterprises employed 99% of a country’s employees, as they did in 1985 (6).  In China, that number was closer to 20%, and basically insignificant in the rural districts.  This allowed Asian gradualism to work:  the government could liberalize some non-SOEs while protecting the SOEs (8).  These countries had large numbers of farm workers who could move to the cities and work for these new private firms, but for most of the EEFSU countries, this large mass of people did not exist.  In fact, Sachs argues, perestroika was the Soviet attempt at gradualism.  But because there was no way that private firms could offer as much as the State jobs offered (at least not at the beginning; in the long run, we know that people are better off in free societies), and because there was no large group of people whose conditions were so terrible that the early private jobs would be a major improvement, the appropriate population for a gradual move did not exist.

So instead of that, eastern European countries realized that they needed “shock therapy,” a term which actually is not as strong as it sounds.  As Sachs notes, the term really just means including SOEs in early rounds of reforms, as opposed to the gradualist method of liberalizing around the SOEs and eventually picking at them.  This shock therapy had the smallest cumulative decline in output, and also could put people back to work in productive enterprises quicker than any other technique (12).

As for democracy, this was not the problem:  the problem was with the authoritarian governments who bankrupted their countries (14).  That is, at least, Sachs’s argument.  I will say that, without the strong constitutional impediments put into place by certain countries (Estonia especially), democratic intrigue did cause harm, particularly for countries which did not make a very quick, strong effort at full-blown shock therapy (Czechoslovakia is a good example here).  Later governments did not have enough public support to go as far as they needed to, and so growth slowed down compared to the booms in countries which could privatize more government activities.

Finally, the Japanese industrial promotion scheme made sense for Japan.  After the war, Japan still had a shortage of raw materials and import restrictions, which limited the ability for their skilled laborers to create products.  As a result, it made sense to promote oil and steel companies there.  For the EEFSU, there was already an overcapacity in the industrial sector (caused by governmental industrial planning), so it would make no sense to increase governmental industrial planning (18).  There are some lessons which could be learned from Japan, though, argues Sachs:  a demilitarized economy, land reform, increased competition, competitive labor laws, private organization, a low budget deficit, and low inflation all helped Japan to rebuild after World War II and become a strong economy.

Rethinking great Madden draft strategies

I wrote a piece on this last year. However, the new season of Madden has caused me to review my strategies. Keep in mind, as the Browns, I run a 3-4 vs. a 4-3, so your mileage may vary. I stand by my offensive slotting from the earlier article, with the possible exception of increasing the QB’s importance. Madden 11 is the first one I’ve played where a bad quarterback can handicap an offense. With Colt McCoy, I felt like I hit a wall where I could go no further performance wise. Sure, I’d score my usual 30+ pts a game, but most of them came from running the ball or defensive scores, the latter of which are almost impossible to predict.

Defense, on the other hand, deserves a look. I’ve got my new rankings below!

1. Defensive tackle — The single biggest difference in this year is how critical a pass rush is, especially for a 3-4. Since you have three guys on the line, one needs to attract a double-team.

2. Pass rusher — Like last year, this is a generic term because for some, it’ll be a defensive end and for others, a linebacker. It’s been true for a season now, but the truth is, linebackers can’t play on the line any more — they’ll just get pancaked. For the 3-4, you probably want this to be a linebacker, since you will almost always blitz one or two. By the same token, a defensive end is a good choice for the 4-3, since you will blitz less frequently.

3. Run stopping defensive end/secondary defensive tackle — Most runs go up the gut in Madden, and this is why you need that second great run stopper (assuming your DT is first). You can probably find these later in the draft, but they’re no less important for being later round picks. Strength and size are key — speed is nice, but not huge.

4. Safeties — These are still your most important DBs. A good safety can hide a bad corner, but not the other way around. They can help in run support and make sneaky good blitzers. Strong and free are about equal, depending on scheme. You usually want a big hitter/coverage guy and a speedy guy.

5. Defensive end — For either style of defense, you need a great D-line. I usually want a pass rusher here too, but I’ve started using three run stoppers on the line and relying on crazy blitzes or linebackers for pressure — it’s worked pretty well. For a 4-3, I’d lean towards two pass rushers, but it’s all up to you.

6. Middle linebacker — He should get a majority of the tackles (or split them with the other MLB), so he needs to be a great tackler first and foremost. If the running back gets this far, you’ll need to stop him quick. For 3-4, I like having one speedier guy for blitzes, but tackling and awareness are more important than almost anything else. Hands are a good tertiary stat because they’ll spy and get the occasional pick.

7. Shutdown corner — You will probably have a hard time finding the very best — every corner in the game usually has some deficiency, whether it’s man to man, zone coverage, speed, tackling, or even awareness. The key is to find the one that works in your scheme. In the 3-4, man-to-man coverage is important, as is awareness and tackling. Since you’ll be blitzing, your corner may have to tackle a running back bouncing to the outside. Here is where a great safety can help.

At this point, you can choose whatever you want, and your team will matter too — the Browns have two awesome corners in Eric Wright and Joe Haden, so I really haven’t spent many draft picks on them. Same with TJ Ward. The biggest change is how hard it is to stop a good passing team without a pass rush. Coverage sacks aren’t very common any more — somebody will get free, and unless you have at least two good to great pass rushers, you’re screwed.

Hey Kevin, the Browns aren’t much better

Of the first three games of the season, I’ve seen one (vs. KC), and I think this is a better team than last year’s. However, I can’t help but think there’s still a lot of work to be done.

1. Play calling — Primarily on offense. This falls on Brian Daboll, the offensive coordinator. In week 1, the Browns barely ran at all. Until week 3, we passed more than we ran, despite the fact that the Browns have a lot less going for them in the passing game.

Also, where’s the Wildcat (we call it the Flash)? You have Seneca Wallace and Josh Cribbs on the same team. We should have teams panicking over this, but it’s almost never called. I counted exactly one Wildcat the entire KC game, and it failed miserably (4 yard loss for Harrison, if I remember right). We did this a lot during the preseason — why not now, you know, when the games count?

2. Defense — I’ve been very pleased by Cleveland’s defense. TJ Ward and Joe Haden have looked awesome. I have no idea what happened to Eric Wright against Baltimore — perhaps Anquan Boldin has compromising pictures of him. Either way, it looks like a one-game blip that he’ll hopefully correct. The pass rush is lack-luster, but this is largely a function of a mediocre defensive line, particularly when Shaun Rogers doesn’t play. If we don’t freak out and go quarter back and don’t pick a wideout, I hope we get an awesome pass rusher.

3. Jerome Harrison — It’s really hard to judge him, since he got 9 carries in week 1, 16 carries against KC, and was out for week 3. However, he doesn’t look very good. Maybe he’s just better as a change of pace back, but either holes aren’t there (doubtful) or he isn’t hitting them (more likely). Weirdly, the Browns call a lot of singleback or shotgun with him, taking Vickers out. Scat-backs generally as much blocking as they can get.

4. Peyton Hillis — Wow. I had no idea this guy could be this good, manhandling Baltimore for almost 200 yds combined rushing and receiving. Clearly, we won this trade from Denver, getting Hillis (and a fifth round pick that could be as high as a third round, I think) for Quinn. He clearly needs to be the feature back, at least until he cools off. He’s more of a straight ahead kind of guy. I’ve heard some comparisons to Mike Alstott. I don’t think that’s perfectly accurate (probably a bit slower), but he’s still the best back on this team.

5. Special teams — Really good kick coverage, but Cribbs has been less of a factor this season. KC almost refused to kick to him.

Overall, the Browns could be 3-0 right now, instead of 0-3. If we can get any semblance of a passing game and Hillis is as awesome as I think he is, I see some wins in Cleveland’s future. I predicted 8 wins in some places — I think we’ll probably get fewer, since our easiest games were the first two, but even if we have the same record as last year (5-11), I think teams will hate playing the Browns this year.

In The Papers: Try To Look More Convincing

Stefano Della Vigna and Matthew Gentzkow have a paper entitled Persuasion:  Empirical Evidence (ungated copy here), in which they look at different forms of persuasion and see where and how it is successful and unsuccessful.


We provide a selective survey of empirical evidence on the effects as well as the drivers of persuasive communication. We consider persuasion directed at consumers, voters, donors, and investors. We organize our review around four questions. First, to what extent does persuasion affect the behavior of each of these groups? Second, what models best capture the response to persuasive communication? In particular, we distinguish information-based models from preference-based models. Third, what are persuaders’ incentives and what limits their ability to distort communications? Finally, what evidence exists on the equilibrium outcomes of persuasion in economics and politics?

The problem with all forms of persuasion is how to get good information from agents who are motivated to make you believe something (3).  For example, what advertising should you believe?  You might think that people who watch a lot of advertisements will have their minds shaped by these persuasive messages, but as it turns out, there is little support for a consistent effect of advertising spending on sales (6).  It seems that people do a good job of tuning out those kinds of messages.  Political ads, especially, don’t change minds (9).  About the best that they do is motivate people who support the candidate already to go out and vote for him.

Get Out The Vote campaigns, however, do work, provided that you do a door-to-door campaign and get real face time with people.  Mailers have almost no effect, and phone calls have a mild effect, but door-to-door work has a fairly significant effect on turnout (11).  Incidentally, television seems to be a culprit in the drop in voting in Congressional races.  The authors speculate that, as television replaced local newspapers, people consumed less local media, and so they did not know as much about their local candidates (13).

One interesting finding in this paper is that payment can have negative effects on outcomes.  For example, women were less likely to donate blood if they were paid $7, although if they had $7 donated to the charity of their choice, they donated blood just as often as if there were no payment.  Men, meanwhile, had no change:  the same proportion donated either way (13).

How about on the markets?  Financial reports are incorporated in stock prices within minutes (15), which is a strong piece of evidence for rational behavior in markets (15).  Interestingly enough, there is actually a longer-term move (on the order of three quarters) with financial reports, which goes against the EMH (16).  Meanwhile, buy and sell recommendations by people on financial shows and in columns do affect market reactions (17).  One problem here is that analysts have a conflict of interest:  optimism is more valuable than accuracy (29-30).  The authors speculate that this might be the case because optimistic analysts become more privy to inside information.  As a result, you are much more likely to see buy or hold recommendations than sell recommendations.  And with financial reports, companies tend to publicize more favorable numbers (28).

The authors look at human actors as Bayesians.  The belief model they come up with predicts persuasion to be more effective if the receiver is less certain of the truth (18).  So if they are unsure of the truth but see an advertisement, this will shift their beliefs.  There are diminishing returns to watching the same ad multiple times, however (19).  But sophisticated agents adjust for the sender’s credibility (21).  Because many ads are non-informative (23), the receiver builds in some level of disbelief.  With the financial markets, there are people who specialize in figuring out what the overly-optimistic analyst says versus what this actually means:  if the guy says hold, you should actually sell.  This form of confirmation bias—an entirely rational response to strategic behavior on the part of the sender—limits the potential of spreading good information (25).  The receiver cannot easily distinguish real signals from opportunistic ones (because the sender has a motive to make both signals the same), and as a result, the receiver discounts all signals to some extent.

So what does help?  Neutral third parties are one way of assisting with signal clarity.  If a third party confirms or denies a claim, this shifts perceptions toward the third-party statement, regardless of the sender’s advertisements (27).  Senders will then try to incorporate these statements into their own advertisements, which is part of why so many automobile advertisements trumpet the awards their vehicle has won (though the award-giving agency in many of these ads isn’t entirely unbiased…).

One final note is with newspaper and television bias.  The authors argue that newspaper slant comes from a demand for bias, not necessarily the owner’s ideology (32).  The authors point out cases in which the same owner runs a conservative paper in one city and a liberal paper in another city.  People want to read articles which comport to their belief systems, and so there is a demand for bias.  I would argue that there is a supply-side issue (especially in the realm of pre-cable television, in which there were a few channels which ran nationally and not enough competition to differentiate the demand) as well, but their point on demand is well-taken.

Poor Choices

Buffalo is screwed for a few reasons.  Yeah, the weather’s cold and it’s not a pretty place to live.  Yeah, the state of New York is in economic Operation Shutdown and the Bills don’t have the big pockets of Manhattanites to count on.  But really, the biggest reason the Bills haven’t had a winning season in a decade is poor talent evaluation and no plan.  Buffalo released Trent Edwards today.  This after deciding two weeks ago that he was their starter and benching him just one week ago.  I’ve spoken my piece on Edwards—Dick Jauron’s mismanagement screwed up what could have been a league-average quarterback—but this is absurd.  They’re planning on bringing back Levi Brown.  Will he come back after being so unceremoniously dumped?  Well, probably.  They want to give Chris Kelsay an extension.  Why, exactly?  Kelsay was a weak left end and is a weak outside linebacker who can’t blitz and is a weak cover linebacker.  Yeah, he was good as a cover guy compared to other defensive ends, but that’s not saying much.

Meanwhile, the offensive line is still a shambles.  Levitre and Wood are good enough (and should develop), but the other three starters suck and their depth is even worse.  Hangartner is a good swing man but a stretch at center.  On the defense, the whole point of a defensive scheme shift is that you go with the system that works best for the players you have.  Buffalo has a lot of undersized defensive players—they have a converted safety (Keith Ellison) playing middle linebacker, and last year had another safety (Scott) playing outside LB—so they go with a 3-4?  I understand that Buffalo’s era of dominance coincided with a 3-4, but frankly, that’s because Bruce Smith was the greatest defensive end ever (Reggie White is a close second, but remember:  Smith did what he did as a 3-4 defensive end, not a 4-3) and there were a lot of very nice surrounding players.  They don’t have that now—they have a good middle linebacker who’s always hurt, a secondary that plays way over its head, and a few knick-nacks at other positions.  So why not play to everyone’s weakness and go to a 3-4!

But what else do you expect from an owner who hires a 72-year-old rookie to be GM?  Even then, I suppose Buddy Nix is still an improvement over that moron Tom Donahoe (with the couple of years of Marv Levy as a relatively successful time, at least when it came to drafting some guys who didn’t suck).  I hate that Buffalo is in such a downward spiral of doom, and probably the only way to fix it is to leave town.  I doubt there will be anybody who would want to keep the team there—even with its history, the Bills don’t do so well sales-wise (go back to the “bad economy” part) and the Toronto games have become a flop—and at this point, new ownership and a guy dedicated to spending 3-4 years cleaning the whole mess from the ground up is probably the only thing that can save this team from itself.  In the meantime, I will continue to play Madden, where those Bills make good personnel decisions and can post winning seasons…