Thread by @rgunns

🚨 ALERT ALERT 🚨 it's time for a new episode of debunking myths about the #GreenNewDeal. today we're talking the recent $93T cost estimate for the GND and why it's a problem.

warning – this one might get a little feisty, but i'm gonna try to keep my nerves under control. 1/

to keep things fun, i'm gonna structure this as a game (since they're clearly playing by releasing this estimate). i'm gonna present 5 doors. behind each is an explanation for why the estimate is flawed. the grand prize is that you never have to think about this again! 2/

WELCOME TO "GUESS THE PROBLEMS WITH THAT GUESSTIMATE!" i am your lovely host, rhiana. think of me as steve harvey but with better gender politics and worse suits.

now, are you buds ready for door #1?! 3/

DOOR 1: YOU CAN'T DO A RELIABLE COST ESTIMATE WITHOUT POLICY + PROGRAM DETAILS.

even in the best scenarios, cost benefit-analyses involve some guesswork. there are lots of reasons for this: needed data could be unavailable; the data might not be granular enough, etc. 4/

and this is the case WITH the major details about policy mechanisms, program administration, and financing spelled out. to try to estimate w/o those details – esp. for a plan as comprehensive as the GND – means that there are a ton of unknown variables. 5/

so analysts doing these estimates have to make a TON of assumptions about how things could be run. (and you know what they say about assumptions...) and that's not always done in bad faith or without rigor! 6/

but it doesn't change the fact that any program in the GND could be structured in multiple ways and each way could incur completely different costs. that means that the AAF estimate of $93T isn't actually an estimate of the cost of the GND as we are developing it. 7/

rather, it's an estimate of what the GND would cost if we structured everything the same way as all of the programs that AAF used to make their assumptions AND that the fed government pays for everything. that is highly unlikely.

that brings me to door #2. 8/

yes, the GND is a federal policy that advocates for a more active and present public sector. but that doesn't mean that the government has to pay for everything! 9/

in lots of areas in the GND – from transitioning the grid to making EVs to making homes more energy efficient – there are large sources of private capital that are ready and willing to be deployed. in fact, in a lot of these areas, business has been WAITING to invest. 10/

but they have been waiting on a clear plan and direction from the fed. government before they invest. the GND gives them that plan AND that opening. that's not to say that private capital will be used for everything or that privatization is even desirable in some areas. 11/

but even still, it is a real option. i mean, most plans about how to transition our power sector assume most costs will come from the private sector (see bit.ly/2SEPfT6). again, not saying that that's the right move, but the money is there and willing. 12/

and even beyond private capital, cost-sharing arrangements between federal, state, and local governments are common and could be used in a GND. obvi, we have no interest in burdening states and communities with unfunded requirements. 13/

still, there are models that could make sense for a GND. for example, the feds could cover the whole cost initially and the rate could drop over time (à la medicaid expansion). the point is we have options. and acting like we don't is a huge issue with the estimate. 14/

DOOR 3: it doesn't deal with redundancies.

because the estimate treats the GND as a laundry list of projects and not a system, it double counts a lot of things and underestimates possible efficiencies. for example, AAF includes estimates for JG and M4A. 15/

about 1/5 of the cost of JG comes from benefits, most of which is related to health insurance. but if you have M4A, the need for insurance disappears and that costs drops, bringing down the cost of a JG. the estimate doesn't take that into account. 16/

similarly, you could use workers in the JG to help build a smart grid, upgrade homes and buildings, etc, which would likely bring down the costs of those projects. but again, the estimate does not consider that. 17/

in fact, certain places in the estimate seem to purposefully make assumptions that increase redundancy and thus inflate the projected costs. in the estimate of the JG, for example, they include multiple scenarios where workers move from underemployment to a JG job. 18/

makes sense – one of the goals of a JG is to pull people out of underemployment and connect them to better-paying careers. BUT the estimate doesn't seem to account for people leaving JG. that means that every person stays in their JG job for TEN YEARS. 19/

is it really realistic? the GND would create tons of jobs – the vast majority in the private sector and the majority of which could likely pay more than JG jobs. the same paper that they cite assumes that JG participation decreases over time. why doesn't the estimate? 20/

(that same report also argues that displacement effects – meaning people leaving the private sector to go to a JG job – will likely be minimal (depending on the structure of the program.) yet, the upper-bound estimate of the JG cost from AAF completely disregards this. 21/

the estimate assumes that every person earning under $30K would move to a JG job. but the average salary for a JG job in the study they cite is $32K. so you mean that everybody is leaving their current job for an extra $2K? 22/

DOOR 4: there is a difference between a *shift* in costs and new spending.

M4A is a shift in costs – meaning that we already pay the same amount in health care costs right now. we just do it through the private market instead of through a single payer system. 23/

thus, the gov't arguably doesn't need to create a new revenue source to pay for M4A. it just has to come up with a way to capture what we already spend. (and the efficiencies created could save much moola.) same with JG. 24/

the fed gov't already supports unemployed + underemployed folks through lots of programs (EITC, SNAP, medicaid, etc.) a JG program would likely reduce costs for all of these programs. 25/

https://twitter.com/rgunns/status/1103070303490662400?s=21