In computing science, a greedy algorithm is one that looks only one step ahead and always takes the easiest next step. Greedy algorithms are simple to implement and almost never optimal for significant problems. Shockingly, they don’t work terribly well for advancing liberty, either. (If you came here expecting a screed on teh ebil Koch brothers and probably something to do with Citizens United, you’re apt to be disappointed. I’m not using “greed” in the Gordon Gekko sense of the word. But stick around anyway.)
A lot of things are illiberal. Taxes are illiberal. Standing armies are illiberal. Financial regulations are illiberal. Many of these things are created seemingly piecemeal, as well-intentioned legislators flail about in the midst of crisis, trying to Make Things Better and believing that they actually can. Since government growth — and corresponding incursion on liberty — tends to ratchet itself ever forward and opportunities to get rid of these illiberal tumours are vanishingly rare, it’s tempting to seize upon every chance that presents itself (hence, “greedy libertarianism”). That’s not such a great idea.
Consider, for example, the Canadian federal consumption tax known as the GST. Originally introduced in 1991 at 7%, the Harper government reduced the GST to 6% in 2005 and again to 5% in 2008. I’ve argued before (and still believe) that taxation is theft, and theft is about as illiberal as it comes, so a reduction in taxation should be powerfully good, right? I certainly believed so at the time.
Turns out, maybe not so much. (See this? This is my deficit hawk face.) As Stephen Gordon explains over at WCI, lowering the GST significantly lowered federal tax income, by about 0.5% of Canadian GDP per year (go look at his graphs, they explain it better than I do). While federal tax income kept rising (albeit slowly) until about 2008, most of this came from offsetting increases in personal income tax revenue. This is… not the best thing if you want to run a strong economy: income taxes (whether personal or corporate) are far more damaging to growth than consumption taxes, and while the latter are notoriously regressive when looked at simple-mindedly it’s pretty well understood that GST rebates can be used to make them, on net, reasonably progressive.
So Ottawa shifted its income base from a growth-neutral GST to growth-negative income taxes right before a major recession. Now we’re running deficits for the first time since 1997 — and as anyone who’s given debt service and the compound interest equation any thought might tell you, deficits now increase the risk of higher taxes later. But fuck it, that’s our kids’ problem, right?
For another example, we look to Niall Ferguson’s narration of the Savings and Loan crisis. (Have I mentioned lately that The Ascent of Money is a great book? So far I’ve extracted three blog posts from it.) S&Ls were originally conceived in the wake of the Great Depression to provide, well, savings and loan (mostly mortgage) services. As originally incorporated, they were limited in the interest rates they could offer (on savings) and charge (on loans), in the investments they could make with their capital, and in the areas in which they could do business. In return for abiding by these limitations, S&Ls’ deposits were federally insured. Remember that, it’s important.
In the late ’70s, S&Ls were getting butchered by both double-digit inflation (which made the “savings” part rather less than profitable) and countervailing interest-rate hikes (which did the same to “loan”). In response, the Carter and Reagan administrations relaxed restrictions on the sorts of things in which S&Ls could invest and on the sorts of deposits they could accept. Suddenly, S&Ls found themselves able to drum up money almost effortlessly and sink it into improbably dubious real-estate investments, while their deposits remained insured. Moral hazard, anyone? You know where that led.
(You can tell a similar story about subprime mortgages and the shadow banking system, by the way. I’d tell that story too, but I’m sick of it.)
The essential problem here isn’t deregulation, as many would like to have it. The problem is haphazard, piecemeal deregulation. One feature of the government-market interface changes (the GST rate drops), but everything else stays the same (federal obligations aren’t reduced). Taking out one block from the tower changes incentives, sometimes wildly. You know how libertarians are always harping on about how sudden government involvement in markets — bailouts, say, or stimulus packages — wildly distorts incentives and leads to more confusion and rent-seeking than actual growth? This is damn close to the same thing.
This is the point in the blog post where I tell you what we should do instead, and leave you to bask in my august wisdom. I haven’t a fucking clue what we should do instead. But it’d sure be nice if policy wonks would take an adversary’s perspective to proposed legislative changes — whether they be the common ratchet of government or the rare opportunity to claw back some liberty — and ask themselves: “How would I break this?”