One of my big concerns, inspired by Chris Papagianis and Jason Delisle among others, is the public sector’s increasing reliance on off-balance-sheet activities to achieve policy objectives. This is one of the reasons why I’m skeptical of balanced budget amendment proposals and setting spending levels to a fixed share of GDP (even though Charles Blahous has made a decent case for doing so). Matt Yglesias explains how setting arbitrary spending limits might lead to the use of crude regulatory tools to achieve the same objectives less efficiently:
Instead of spending money on food stamps, you could require supermarkets to give discounts to low-income families. Instead of taxing gasoline, you could have ever-stricter fuel efficiency standards. And the instead of using the revenue to finance deployment of clean energy technology, you could have a regulatory mandate on utilities to do it. We already do a fair amount of this kind of thing because it makes it easier to hide the ball in terms of who bears the costs. But because these policies are less transparent, they are much more prone to perverse, unintended consequences. Thoughtful people should be trying to move politicians off this incestuous blend of corporate paternalism and overregulation, and onto the sweet terrain of paying for services with taxes. A constitutional rule barring spending in excess of 18 percent of GDP does just the reverse.
Indeed, Adam Ozimek suggests that minimum wage legislation is best understood as this kind of a cost-shifting intervention:
What the minimum wage does is effectively push the costs of a multi-billion dollar, illusory, anti-poverty program onto employers. It’s an inefficient way to help poor people, and a hidden tax on businsses that forces them to spend $4 so that the government doesn’t have to spend $1.
More broadly, conservatives overestimate the importance of shrinking explicit spending relative to the cause of shrinking shadow expenditures, like credit guarantees. Sometimes shrinking the latter — which are in many respects more dangerous than the former — will mean the former will have to increase.