Author Archives: natecarroll

About natecarroll

I'm not a biter, I'm a writer.

No, Noodles & Co. Did Not “Punk” Anybody

Fast-casual dining spot Noodles & Co. IPO’ed on Friday for the price of $18 per share, and immediately popped by nearly 100%, eventually closing the day at $36.75. It has since gained another 30% and currently sells at $47.25. Obviously, some underwriters fucked up.

Somewhere beneath all the facts and data is an interesting, accurate story about why they fucked up, but that’s definitely not what Daniel Gross wrote for the Daily Beast today. Instead, he explains that the food snob Wall Street-types were so blinded by the bubble of NYC cuisine, they missed the appeal of Noodles & Co. simply because of its Midwestern focus.

Plainly, Wall Street blew it. The shares soared not because there is some bubble raging in restaurant stocks, or because noodles are a hot new concept. Rather, they soared because traders outside the bubble of Wall Street and the institutional investors they deal with had a different view of Noodles & Co.  And when you take a step back, it seems as if Noodles & Co. was actually set up to punk Wall Street.

I’ll give Gross the benefit of the doubt and assume he was just doing his best to write an entertaining story. It is an entertaining story! But it fails as an explanation of what’s actually going on.

While there probably is some truth to the assertion that Wall Street bankers are snobbier than the average Midwesterner, they’re not stupid. One of any number of cliches could be inserted here: numbers don’t lie, money talks, etc. The slick alpha-types at Morgan Stanley and UBS could surely see past our inherent dowdiness if doing so meant however many extra millions of profit.

Eric Cantor is Right About Obama’s Climate Plan

Although his Twitter buzz is currently getting overshadowed by the SCOTUS ruling on the Voting Rights Act, Obama started his big push for action on climate change today. As expected, the administration is sticking to its ‘more of everything’ approach, which means the Climate Action Plan released today is a sprawling document that leaves a lot of questions about how the vague initiatives will all play out.

Helpfully, Eric Cantor offers his summary:

@GOPLeader: More taxes and regulation on America’s energy suppliers will only make energy prices go up.

This rebuttal is both devastating and stupid, because the whole point is to make energy more expensive. Burning fossil fuels exerts a cost on the environment that is not reflected in energy prices. All policies designed to address this problem should follow the same fundamental mechanism: they will reduce national energy usage by making it more expensive.

In the long-term, investments in renewable energy might pay off by bringing $/gigawatt prices below the current levels set by low-cost, high-pollution producers. But there’s short term pain associated with those investments. This is true whether taxpaying Americans are paying for the R&D costs directly, through federal grants and subsidies; or indirectly, through taxes on greenhouse gases that will spur private-sector investment in the green alternatives that suddenly have a chance at being the cheapest option.

A real argument might question whether the economy of a very large, spread-out country used to powering itself with cheap energy can withstand the economic headwinds of a large increase in fuel and electricity costs. Especially if that country is still trying to recover from a particularly long and deep recession. Given the support for borrowing from the future to stimulate our way out of economic malaise through federal spending, should we not also consider “borrowing” some degree of environmental health in order to leave a stronger economy to future generations?

I’m not sure of the answer. But it’s a fair question, and one that I wish conservatives would make more explicitly, rather than ignoring or outright denying the effects of climate change.

#1 Way to Reduce Uncertainty at the Fed

James Bullard, the President of the St. Louis Fed, recently gave an interview to Wonkblog explaining his decision to dissent at the Fed meeting last week.  The one-word summary would be “inflation,” but his answers as a whole describe a coherent strategy that’s quite a bit different from what the Fed’s actually doing. Bullard maintains a singular focus on data, rejecting the use of calendar targets as guidance for Fed policy. He observes rightly that his colleagues are half-assing their commitments on this front:

There’s some effort to make that a little more state-contingent by talking about 7 percent unemployment as a threshold. But I’m not sure the committee has thought that through completely, because my own forecast has unemployment continuing to drift downward as the year goes on, so that could pull up the tightening sequence faster than the committee really wants to. I’m not sure we’ve completely thought through that threshold issue. And there’s no reference to the issue of what inflation is doing at that point.

The  whole interview made me question why Fed policy isn’t just determined algorithmically. Assume for a second that the committee could agree on a formula based on unemployment and inflation indicators that would determine rates and purchases. This would not only prevent them from setting policy based on overoptimistic timelines, but it would also reduce the oft-cited problem of “market uncertainty.” Under the status quo, everything Bernanke says is parsed and dissected every which way, with the markets often swinging wildly as a result. The vagaries of the English language are being subjected to more pressure than they can handle.

As a starting point, Evan Soltas throws out a couple of mini-algorithms in his article discussing Wall Street’s mistrust of the Fed.

For instance, it could promise to keep policy accommodative until the economy grows at an annual rate of more than 2.5 percent and monthly growth in payrolls exceeds 200,000. That would convince the markets that bad economic results will extend the monetary stimulus — and make the promise Bernanke has already made more credible.

Unlike most other game-changing economic ideas, there’s no political barrier standing in the way of this one. The FOMC could declare today that all future policy will be set by a public algorithm, and since the Fed was designed to be free from direct interference by other branches of government, no one could do anything about it.

Randomness as a Cure for the Most-Popular Clique

Over at The AtlanticDerek Thompson has a good post breaking down the feedback loop created by “Most Popular” lists. Pulling no punches, he describes the effect as tyranny, and he backs up his assertion with inside info on click rates at The Atlantic web domain. The bottom line? No matter how much design/SEO expertise has gone into the web pages there, the most clicked links are the ones in the “Most Popular” sidebar box. The tautology should be obvious: the most read articles are the ones that are read the most.

But the point of Thompson’s piece—and where the tyranny comes in—is that the interaction between the “best” and “most popular” dimensions flows both ways. That is, perceptions of popularity can actually change the way we experience something.  The effect has been experimentally isolated by a pair of sociologists who created a faux online music market, then twiddled with the supposed popularity rankings of the songs and observed the changes in consumer downloading behavior. The least popular songs of the control condition  became had-to-have downloads when participants were instead told they’d been the most popular with previous participants.

Thompson sees this effect as disappointing evidence against self-determination, writing, “When we outsource our navigation of the world to other peoples’ opinions, we lose, in a small way, our ability to individually evaluate the quality of our experience.” But, realistically, there’s way too much stuff in the world—and, in particular, on the internet. People need a quick and easy way to decide what’s worth reading, watching, or listening to, and popularity acts as a mostly-suitable proxy for quality. From an individual standpoint, consumers that follow the herd are just behaving rationally.

The popularity principle becomes disturbing only in aggregate, when the resulting positive feedback leads to a winner take all environment where great content can fall by the wayside if it’s not immediately discovered, and perfectly average media can rise to the top overnight. Think of the recent mega-success of “Fifty Shades of Grey”: it’s not necessarily any better than the other hundreds of erotic novels written by anonymous nobodies (and it’s certainly not thousands of times better, as a linear analysis of sales figures might lead someone to believe), but a confluence of factors brought it onto the public’s radar, and the “most popular” effect took off from there.

Facebook and perpetual wannabe Google+ would like to solve this problem by tabulating popularity within the confines of a self-selected group rather than the overall population. Although this reduces the winner-take-all effect, it tends to cut people off from opposing viewpoints by enclosing them in an echo chamber of their own choosing. One only needs to spend a few hours on Reddit to realize that similar self-selecting mechanisms create a weird insularity that undercuts merit evaluations.

A mobile app developed recently by a U.K. studio gets right to the heart of what’s needed to keep the online mediascape fresh: it’s a photo-sharing program called Rando. Billed as an “anti-social” app, Rando eschews nearly all of the functions of its competitors; pretty much all you can do is send a picture to a random user somewhere in the world, and then wait to receive one in turn.

Rando might be excessively minimalist, but it breaks resoundingly from the online obsession with popularity. As one of the co-founders told Tech Crunch, “We want the emphasis of taking a photograph to be less deliberate in the sense that users aren’t driven to share a picture purely because of how many likes/ comments it will receive.” He’s talking about freedom on the producer side, but it’s equally freeing as a media consumer to not have to worry about who to follow or what pictures to click on. The tyranny is lifted.

The conception of randomness as freeing is also why I think a simple service like Pandora can be more usable than an unlimited download subscription such as Zune Music. From a rational perspective, consumers are never worse off with more choices, but in reality, downloaders get overwhelmed and defer to the popularity bubble. Other online media companies would do well to inject some randomness into their presentation schemes. For a specific example, Netflix ought to create “channels” that let viewers tune in to a mix of TV episode or shorts.

Gentlemen Prefer Bonds

Via Evan Soltas on Twitter, here is an interesting paper making the case for pollution bonds as a more socially efficient way to tax negative climate externalities than a straight carbon tax.

The regulatory scheme would require emitters to purchase bonds with a face value equal to the highest reasonable estimate of the social cost of emissions. At maturity (30 years, probably), these will pay out the difference between the bonds’ face value and the most up-to-date estimate of the costs of emissions discounted to the date of issue.

Without a resale market, this is equivalent to a Pigouvian tax stretched out across decades to account for the fact that estimates of the true costs of emissions will be better in 2043 than they are now. But if the bonds are tradable, the uncertainty can be transferred from the emitter to whomever wants to buy it. The paper notes that market mechanisms usually produce better predictions than systems based on data analysis, citing commodity futures and the Iowa Electronic Markets as two noteworthy examples.

In general I’m a fan of using prediction markets to guide policy (see everything that Robin Hanson’s ever written for more on this issue), but I’m not sure it would work in this case. The problem here is that future outcomes are “messy” in a way that’s not usually the case with election outcomes or box office receipts or any of the other categories Intrade used to let people trade. The whole point of the scheme is that we don’t know how to measure the true social costs of gas emissions—but to resolve the bonds, some regulatory body is going to have to do exactly that. This means the traders aren’t betting on the true costs at all; they’re betting on what some government muckety-mucks will declare the true costs to be.

If there’s enough time between a bond sale and maturity, traders with superior knowledge of the true costs might use it as a benchmark for the value of the bonds, but they’re counting on the regulator to catch up at some point before the bonds expire. Either way, the tax to polluters is going to be based on some government agency’s best estimate of the total social cost of emissions.

My main takeaway is that polluting should be a lot more costly than it is now, regardless of what it takes to make that happen.

Obama’s Voluntary Pay Cut

Matt Yglesias just put up an odd little post about how absurd it is that the president is giving back 5% of his income ($20,000/yr) directly to the U.S. Treasury in a supposed show of solidarity with federal employees facing furloughs and pay cuts as a result of the sequester. I agree with this conclusion, but his reasoning is surprising:

Here’s my advice. If you feel like you’ve got too much money on your hands, don’t cut your own pay. Take the money, walk down the street, and give it to a homeless person. Or put cash in envelops and mail it to random families. Or get fancy about it and use GiveDirectly to fight poverty in Kenya. But do something! Money is valuable, if not to you then to someone else. Use it wisely.

The implication here is that giving money to the government is akin to throwing it in to a black hole; you’d be better off just handing it out at random. There are, of course, people that subscribe to this viewpoint, but most of them are in the Tea Party.

The U.S. government is trillions of dollars in debt and will be facing budget deficits for, at very minimum, the next 10 years. Although the consensus view from economists and thinking commentators is that our debt load is manageable and doesn’t deserve the “crisis” label many have tried to put to it, there’s no question Treasury could use the money.  Relative to the number of digits in most federal budget figures, it might seem like Obama’s extra contribution just disappears, but on the margin it’s making an impact. Anyone who believes the U.S. government is a system that deserves to be funded (hint: you, me, and Matt Yglesias) should be happy to see revenues going up.

My problem with the gesture is that it’s not much of a sacrifice. Obama and his family will clearly survive just fine on the other $380,000 of his salary and the state-provided digs–not to mention the speaking fees once he leaves office. $20,000/year is a lot in absolute terms, but it feels insincere for someone so well-off to act like he shares the struggles of the common man, some of whom are facing cuts of much more than 5%. It’s just an act of political theater, and one that’s drawing away time and attention from substantive issues.

QE in Japan

The Bank of Japan announced on Thursday a plan to expand the money supply through significant and sustained long-term bond purchases. This step was more or less expected as part of Japan’s new, more accommodative economic plan, the so-called Abenomics–following from the name of the recently elected Prime Minister, Shinzo Abe.

In contrast to the austerity policies spreading throughout the EU–which don’t seem to be working–Abe came into office with a plan for big stimulus spending.  Japan has been plagued by deflation for almost two decades, and the economic woes are certainly apparent in a historical chart of the Nikkei 225:

nikkei-history

As in the U.S., the zero bound for overnight interest rates has constrained the central bank, and now Japan is following our lead by pursuing a strategy of quantitative  easing. In fact, buoyed by the relative success of QE here in the States, Japan will be doing it bigger. As the FT reports:

Under the new measures, the BoJ will expand its balance sheet by 1 per cent of gross domestic product each month this year and by 1.1 per cent per month in 2014, according to estimates from Barclays.

By comparison, the US Federal Reserve’s current monetary easing programme involves increasing the balance sheet by 0.54 per cent of GDP per month.

“This is really taking policy where it hasn’t been before”, said Jonathan Cavenagh, senior FX strategist at Westpac in Singapore. “It’s pretty bold. It has certainly taken us by surprise.”

The Nikkei gained about 2.2% on the news, but it remains to be seen how the policy turns out on the whole. Either way, it will be good to have some more comparative data to help determine the effects of QE on developed economies.