Unemployment That Doesn't Suck: Part 3 - How We Might Fix It

A better combination of incentives, and additional transparency about what they are, could go a long way in making the system work much better for many more folks

If you haven’t read Parts 1 & 2, Part 3 won’t make sense. Catch up here :)

---

What’s Not Happening

When I started this deep-dive, I expected to find some kind of smoking gun. Something very obviously wrong with the system that would be addressable though policy action. I didn’t find that. No one seems to be acting in bad faith here, the system is just bogged down by the kinds of standard misalignment and opacity common in any public-private partnership.

The good news there is that no one is really directly out to get us here. The bad news is that means fixing it is much harder to do.

Misaligned Incentives

The core issue that I see in this “market” is that the sides of the network aren’t aligned in the incentives:

• “Users” want timely payments in the format that works best for them to use most effectively

• The State wants to provide those payments as efficiently as possible

• The Provider wants Users and the State to choose the payment method that is most economically advantageous for them

Users and the State want the same thing, so long as Users don’t want paper checks / actual physical cash. On-time payments that are easy to send/receive. Obviously some Users will want those things, but I think we can all agree that in 2023 that’s ground we can cede.

Structural reasons folks might prefer physical options exist and suck and should be addressed and I’m not going there today.

The misalignment, then, is with the Provider (Conduent in this case). Conduent only makes money when people receive and use their EPC. If that’s not what works best for Users, they don’t really care. And they stack the deck in favor of their own offer so that the State thinks it’s the most economical option as well. It’s the Users who get squeezed here.

Opaque Structure

That last paragraph might be completely wrong. I cannot find any documentation on how / if Conduent is involved in processing payments not destined for their own managed EPCs. I cannot find any documentation on exactly how the State pays them. I don’t know what the contract terms are generally, like how long it lasts or what the bidding process is for potential replacements. I don’t know how much the State believes the current structure saves them vs. the previous implementation or other alternatives.

I don’t know anything other than Conduent provides a service to the State and that the default relationship is the one that makes them the most money. For a public-private partnership, the details are awfully private. That is, unfortunately, all too common and it makes any kind of policy recommendations extremely difficult to make stick.

That’s intentional, and it’s bad government.

A Better Path Forward

With that caveat, I still think there are recommendations that would improve the situation for unemployment payments and numerous other similar programs at both State and Federal levels.

Pay on Program Size

Payments from the State to Providers should be based purely on the number of Users active in program the Provider is servicing. In this case, that means that the State would pay Conduent based only on the number of unemployment claims processed each period, regardless of the fulfillment method. This would align the incentives across the board; Users choose the payment method that works best for them, the Provider acts on that choice, and they’re paid by the State.

No need to push a suboptimal experience, just make sure the full program is working as efficiently as possible. If that means less EPC and more Direct Deposit, no one will be acting against that.

Published Payment Schedules

We know the number and amount of unemployment claims processed - that’s always published. “Fulfillment costs” to the State should be published as well, so that taxpayers get full transparency to the true cost of the program. This would also increase transparency across States / Programs, where any outlier could be audited and an alternative could be explored. If everyone in California knows how much more they’re paying for the same thing relative to Washington, it’s much easier to push for a change (and be specific in that push).

No Default Options

I never actively chose to receive an EPC for my unemployment payments. I doubt anyone does; it’s the default option that you must opt out of. There’s no reason to have a default option here once the incentives are aligned. And even in the misaligned current-state, these are actively unfriendly at best and shady at worst. They shouldn’t exist - Users should be “forced” to make a choice and the State and Provider need to live with that market outcome.

Yes, I know - this is all a pipe dream and it’s really not how American bureaucracy works. But, I think we can and should get closer. Especially when solutions are obvious and available.