I of course endorse Natalie Avery’s case for making the liquor license approval process more collaborative and less adversarial. There are some cases where adversarial regulation is really appropriate. Something like Spencer Bachus’ view “that Washington and the regulators are there to serve the banks” is totally inappropriate for that sector.

But local government should want to see businesses grow and prosper. Where the need for regulation arises, the goal should be help firms comply with the regulations not be super-adversarial about it.

In general, I think we don’t spend enough time thinking about when the “nice regulator” model is appropriate and when the “mean regulator” is what you want. But I think the main question to ask yourself is “would deliberately violating this rule on a consistent basis be a smart business strategy?” Think about restaurant kitchen sanitation. All else being equal, if you own a restaurant it’s in your interest not to make your customers sick. Thanks to information asymmetries, I don’t think we can trust this entirely to the market. Regulatory inspections are a fine idea. But we don’t need to worry about restauranteurs willfully trying to run unsanitary kitchens. The regulatory agency here should be a “nice regulatory” model that tries to help businesses understand what they need to do to get up to code.

Banking’s not like that. It makes a ton of sense for bankers to invest lots of time and energy in deliberately locating and exploiting loopholes and the like. Here you need a “mean regulator” who’s suspicious of the motives of regulated entities and always on the lookout for problems.

Paul Krugman and Kevin Drum recently wrote about the problematic definition of “money” in the modern world.

I think this issue can be helpfully illuminating by dusting off one’s BA in philosophy and attempting a little metaphysical analysis that will help clarify what the actual issue is here. What is money? Well money is currency. And it’s easy to say what the currency of the United States of America is: dollars. So what’s a dollar? Well the word is ambiguous. But a dollar is a unit of account—you can give the price of things that aren’t dollars in terms of dollars. And a dollar is also a medium of exchange. One dollar, is a perfectly safe perfect liquid asset with a value of $1. Four quarters are a perfect safe investment in dollars. Four quarters are worth one dollar by definition so they can never lose value in dollar terms. And they’re perfectly liquid: as long as you’re in the USA, anyone will accept four quarters in exchange for goods or services valued at $1.

The problem of the “broader aggregates” is that there are lots of things that have properties closely resembling those of dollars. My checking account with PNC Bank is basically perfectly safe (thanks to the FDIC) and it’s almost as liquid as quarters. The vast majority of stores will accept my debit card and there are ATM machines all over the place where I can exchange electronic checking account commitments for physical dollars. Something to note here is that some forms of physical currency actually fall as short or shorter of ideal-type money than checking accounts. A sack containing 10,000 Sacagawea dollars is, in most situations, a less liquid asset than an ATM/debit card linked to a bank account worth $10,000.

But circumstances can change rapidly. I remember a few years ago I was in New York during a giant blackout. That caused everyone’s ATM cards to be suddenly, albeit temporarily, useless. 999 days out of 1,000 the guy with the sack of Sacagawea coins is a moron, but when the power goes out he’s everyone’s best friend. The impact of this massive monetary shock on the city’s economy becomes difficult to evaluate since obviously the general lack of electricity was a substantial shock that caused all kinds of asset price fluctuations (perishable food plummets, batteries and flashlights skyrocket) and rendered many capital goods useless. But it’s interesting to try to think about.

What if there was no blackout, but for some reason all ATM/debit cards in the Denver area didn’t work for a week. Posit that the nature of the problem was well-understood and everyone is perfectly confident that their accounts will work again in seven days. So it’s not a banking panic, it’s just that suddenly these assets turn temporarily illiquid. The answer is you’d have a giant, albeit temporary, collapse in real output. Households would immediately pare back spending in order to conserve cash. Business relationships that are embedded in continuing social relationships would continue on a credit basis, but nobody would make any large purchases from strangers. Mass layoffs would be avoided only because the duration of the monetary crunch was known to be temporary and no boss would want to act like a giant asshole about the problem. Having the Army fly helicopters over the Denver area and literally drop dollars on the city would in fact boost growth.

So to return to where we started, the real issue Krugman is pointing to is not difficulty in defining money, it’s difficulty in quantifying the money supply. That’s primarily because the money-ness of different things can change over time. If money-ness were a constant property of assets, then you could say “a checking account is 99% money-like, so we’ll say a $10,000 checking account constitutes $9,900 worth of money supply.” The blackout, however, doesn’t eliminate the checking out instead it makes it (temporarily) much less money-like. And it’s challenging to quantify in real time how money-like different things are. We can say qualitatively that the “shadow banking” panic started to render large classes of assets less money-like. Letting money market funds go bust, as some retroactive bailout opponents think we should have done, would have had the same impact. So do radical revisions in estimates of the safety of short-term Italian debt.