...Archive for January 2021

Fix the Senate I: Scrap the filibuster

The United States Senate is a catastrophe.

It is prima facie anti-democratic: 38 million voters in small states have 15 times the representation as California’s 40 million voters. Extra-constitutional Senate rules and practices, most notably the maximalist filibuster, have turned the body into a “kill switch“, one veto point to rule them all. They have enabled a minority faction to block and obstruct, to prevent any constructive governance at all except when it will redound to the political benefit of that faction. Institutions of government matter. “Democracy” is not some free-standing good thing. If it is to be good, and if it is to survive, it must be embodied in particular, functional institutions. In the contemporary United States, that is not the case.

The Senate is hard to fix. Disproportionate representation of citizens in the Senate (described as “equal Suffrage” of states) is specifically excluded from amendment by the Constitution. In this series, I’ll describe three reforms that I think would improve the Senate, ordered from least to most speculative. The first is easy to enact and very widely discussed. I’ll just add my voice to the chorus.

Right now, this week, we should improve the Senate by eliminating the filibuster. Ryan Cooper has an excellent piece on how this would work (with a pointer to Akhil Reed Amar, whom I love for his work on lottery voting).

I’ve come to oppose the filibuster reluctantly and with some sadness. In a better-arranged polity, modest supermajority requirements might play a useful role. A democracy at its healthiest governs by persuading a broad center, rather than lurching back and forth between irreconcilable agendas of hostile but nearly balanced factions. In defiance of the median-voter-theorem, tack-to-the-center claims made on its behalf, the United States’ two-party system, has reshaped us into two hostile factions, nearly balanced not in absolute numbers but in terms of how our system translates numbers into power. We are, as Lee Drutman puts it, caught in a “doom loop” that is entirely an artifact of a bad political system, not the existential struggle between distinct peoples too many of us are lazily permitting themselves to conceive. Simple majoritarianism is a cause of this catastrophe, as it tempts people to govern with 50% + 1 coalitions that entirely ignore the values and interest of the other 50%. This is an ugly kind of polity. However, the Senate’s effective supermajority requirement via the filibuster has not overcome the incentives of the two political parties, and the incumbents they run, to divide us. It has merely incapacitated us. The filibuster has become a minority veto, which, given the disproportionality of representation within the Senate, could in theory be exercised by Senators elected by fewer than 10% of voters. In practice, it’s not quite that bad, but still terrible. Under current circumstances, a blocking coalition of Republicans can be mustered from Senators representing only 25% of the population, elected by a share of voters even smaller than that. (Even in red states, Republican senators occasionally receive less than 100% of the vote.)

There is a case to be made for a supermajority legislative threshold. But an effective 75% threshold in terms of population represented (ignoring incomplete support among voters) is a prescription for paralysis. Worse, the threshold is asymmetric. While Democrats face that 75% threshold, Republican-backed initiatives pass with assent of representatives of only 54% of citizens. This is an institutional embodiment of “my way or the highway” for the Republican Party. Both parties can force inaction, only one can enable it, despite approximately balanced support within the population. [*]

A principled supermajority requirement might be established in the House, rather than the Senate, in combination with redistricting or other reforms that ensure representation in the House is proportionate to voter support. The current Senate filibuster is simply indefensible. The Democratic majority should scrap it, today.


[*] A bit of an irony that I’ve not seen discussed is that in 2020, Democrats gained control of the Senate, but actually lost the chamber’s overall popular vote to Republicans by a slim margin. (The numbers cited exclude the George special election results, but they’d not change the basic picture.) In 2018 and 2016, however, Democrats won more Senate votes than Republicans, by substantial margins, but did not win control the chamber. Obviously, the US Senate’s composition during any given Congress reflects the results of three elections, so even if we adopted a voting system that made representation in the chamber proportional by party to overall votes received, a given cycle’s winners would not necessarily gain control. However it’s a bit weird and wacky that Democrats gained seats and control during a cycle when they lost the popular vote, while in 2018 Democrats lost seats despite a blowout popular vote win. The quirks in our system can cut both ways.

Missing the forest

Atrios does a good job of capturing how I actually feel about people on the so-called center-left (most notably Larry Summers) wonkifying their way out of support for near-universal $2000 checks. They are mistaken on the narrow technocratic grounds over which they claim authority, an authority that decades of experience prove they absolutely do not deserve. But they are mistaken (or worse than mistaken) at a more fundamental level. They do not get (or if they do get, they have never deigned to address) that the point of universalist benefits isn’t to stimulate the economy, nor to maximize utility while minimizing a fiscal outlay, but to reorder the relationship between citizens, the state, and one another.

Let’s start with the most straightforward case against $2000 checks. Here’s Catherine Rampell:

Simply put, sending money to nearly every American family to ensure that help gets to the much tinier fraction who actually need it is not a terribly efficient use of resources. The payments end up being a pittance for higher-income, fully employed households, yet insufficient for the households that suffered large income losses. The error is compounded if these funds come at the expense of more targeted relief measures — such as expanded unemployment benefits, which the new law guarantees only through March.

Advocates on the left argue that there is room to do both. But even if you believe that there are no real debt constraints, given how low interest rates are, there are still political constraints. Republicans set a $900 billion cap on relief measures in the final deal, which meant that funding stimulus payments required shortchanging unemployment benefits (and state and local assistance).

First, it’s worth noting that Rampell is misleading on the facts here to help buttress her case. It is true that Republicans initially tried to cap the deal size, and so that when $600 checks were included in the current version of the relief bill, the funds were clawed away from other uses. However, the cost of a potential increase from $600 to $2000 was never proposed to be taken from elsewhere. What Republicans like Donald Trump and Josh Hawley proposed was to simply increase the size of the bill.

Rampell is taking a constraint as given, immutable if not by hard economics then by hardball politics, even when by the time she wrote, the constraint was already contested within the political caucus supposedly imposing it. But political constraints are endogenous, and the very purpose of universalism is to reshape them. It blows my mind that Rampell quotes an estimate that the proposed checks would benefit 94% of US households like that is a bad thing. If you believe (as I think Rampell does!) that a just and functional polity will require a larger state footprint in fiscal affairs (which need not not imply bigger deficits, but larger gross outlays which may or may not be matched with offsetting taxes) then surely the program’s benefiting lots and lots of people should be scored a plus!

If you imagine you are the wise planner, given $400B-ish to spend as you choose, then yes, maybe you could do more good with some allocation other than universal-ish checks. But who is living in political fantasyland in this thought experiment? The $400B only exists as potentially allocable money because universal outlays are popular. And a state that typically provides benefits on universal terms will be popular in general, relaxing constraints on further state action — on outlays, but also on progressive taxation, which the state’s beneficence will render more palatable. And progressive taxation will help target net benefits, despite untargeted outlays, relieving economic (as opposed to political) constraints on state action, while also countering democracy-corroding wealth concentration and improving the predistribution. It’s win-win-win!

Paul Krugman writes

No binding budget constraint for the feds, so this is all about politics. And my sense is that broad issuance of checks is actually kind of a loss leader, helping to sell a package that includes UI

At the time he wrote this, still-too-stingy UI had already been passed and signed into law as part of the relief package, and no further expansion was on any political horizon. So maybe original the $600 was the sugar for a broader public to help make more desperate medicine go down, but, as with Rampell, there’s this reluctance to acknowledge that universal benefits might be worthwhile on their own terms because the space of the politically possible expands when broadly everyone is a beneficiary. An increase now from $600 to $2000 wouldn’t be a “loss leader” for any other thing. It would, however, increase the public’s satisfaction with and broad enthusiasm for government and therefore increase the probability that America rediscover’s the virtues of collective action in some broadly general interest, without which we face an apocalypse.

Technocratic authority so often just gives cover to myopia. It’s easy enough to compare different allocations of a hypothetical $400B, and sound smart dissing the “untargeted” one as suboptimal. It can be gratifyingly technical to compute (under tendentious assumptions) little cost-benefit analyses, one controversy at a time, taking the horizons of any deeper future as given, immutable as a lazy shorthand for unknowable. That is a way of sounding smart while being stupid. It has gotten us just where we are.

Judged not as a one-off, but as precedent and practice, universalism is good policy on technocratic grounds. Myopic optimization in response to every shock scrambles incentives (as people try to account for ways their actions might place them in or remove them from classes that will seem worthy of assistance). It creates embittered losers every time, people who feel deserving of aid but whom policymakers decide (optimally or arbitrarily) to exclude. In practice ad hoc optimization is gamed (see, for example, the contemporary PPP program), leading to widespread perception, and also the reality, of corruption, further discrediting government action. The Obama-ites thought their financial crisis response was “smart”, putting out fires while putting wooly concerns about justice and power to the side. It was optimal in the same way that the next dose of heroin is optimal for an addict. It’s true, after all, that absent a dose, the addict will get very sick! We need a theory of state action that is effective at addressing current crises — even profound crises like COVID and the 2008 financial collapse — while supporting rather than undermining social cohesion and the legitimacy of the state. Universalism is that theory.

In the previous post, I wondered: If we now conceive of racism as systemic and institutional, rather than individual and personal, could we also do the same for love? Universalism is one way of embedding unconditional love in the shape of our institutions. The warmth of our civilization flows to all, not in the form of threadbare assistance to those who prove they “deserve” it, but as benefits that we all share. It’s the contribution to this social foundation that we’d means-test, through the tax system. One can imagine, neoliberally, that taxation to support a generous universal foundation would so blunt incentives to produce that we’d collapse to destitution. Have you noticed, neoliberally, that we are already in collapse due to precarity, social resentment, incapacity to manage conflicts and coordinate through government? Is it really inadequate incentives to succeed that are holding back our advances, technological and otherwise, or the fact foregoing a career at McKinsey or Google to do something interesting is just too risky, given how precariously we all live, how far there is to fall? Universal benefits are popular because they are actually good, in both a technocratic and ethical sense. They are opposed by incumbent interests not because they would fail, but because they might succeed.

I’ll let Atrios have the last word:

[T]hey would prefer plunging the economy into a deeper recession and the misery of millions of people on the off chance people might realize government is actually capable of doing things for them.