School districts in the US watch over and educate kids starting at age 5. The current arrangement has a number of gaps, including after-school care, summers, and ages 0 to 4. The Democrats have said they will fill the latter gap in the upcoming reconciliation legislation by providing universal pre-k for ages 3 and 4 and subsidized child care for ages 0, 1, and 2. In this piece, I outline the relevant details of these two programs with a primary focus on their problems.

Universal Pre-Kindergarten

The universal pre-k program is the simplest family benefit program in the reconciliation bill by a large margin. Under the program, states will submit plans to the Treasury for establishing state-level pre-k systems that provide free services to all children regardless of their family’s economic circumstances. Once a plan is submitted that conforms with the various requirements of the bill, the Treasury will send the state some cash so that they can fund their pre-k system.

In the first three years of the program, the federal government is pledging to cover 100 percent of the costs of these pre-k systems. By 2028, the percentage drops to 60 percent, meaning that states will have to find a way to finance the remaining 40 percent from state-level taxes.

YearPercent of Pre-K Expenses Covered Federally

State taxes are generally more regressive than federal taxes and so it would be better to fund the benefit federally rather than on the state level.

More importantly, this kind of state-level cost-sharing is likely to result in many Republican-led states refusing to participate in the program.

Democrats used this exact same structure in 2010 when they enacted an expansion of the Medicaid program. In the first years of the program, the federal government covered 100 percent of the expansion costs, but, in later years, the number dropped to 90 percent. Over a decade later, there are still 12 states that have refused to participate in the program.

The Medicaid expansion problem has been so persistent that, in this very same reconciliation bill, Democrats are proposing to fully fund Medicaid expansion in the 12 non-participating states.

States that balk at covering 10 percent of the Medicaid expansion are likely to also balk at covering 40 percent of universal pre-kindergarten. But Democrats seem dead set on making the exact same mistake as they made in their last piece of signature legislation.

Subsidized Child Care

Insofar as we currently cover ages 5 through 18 with free, universal schooling, and Democrats are proposing cover ages 3 and 4 with free, universal pre-k, you might think that the natural solution for ages 0, 1, and 2 would be free, universal child care. But you would be wrong. When it comes to this age group, Democrats go back to indulging their taste for byzantine means-testing spiced up with a few exotic flourishes like a $1 million asset test and an “activity” requirement for the parents of children receiving child care services.

The basic financing structure of the child care proposal is similar to the universal pre-k proposal. States that want to participate in the program will write plans that they submit to the Treasury who will then provide cash to the states. The financing in the pre-k proposal is open-ended and set equal to a percentage of the state’s pre-k expenditures. For child care, the first 3 years are funded by a fixed pot of money established for the entire country. Starting in year 2025, the funding is open-ended and set equal to 90 percent of each state’s child care expenditures.

YearFederal Contribution
2022$20 billion for entire country
2023$30 billion for entire country
2024$40 billion for entire country
202590% of state child care expenditures

For years 2022 through 2024, it’s quite possible that the Treasury runs out of money and is unable to fund all of the state plans submitted to it. Even if you fill in these three dollar values with numbers based on an estimate of what it would actually cost to cover benefits for these three years, the Treasury may still run out of money as those estimates could easily be wrong.

For year 2025 and after, the Treasury has an open-ended bucket of money for the child care subsidies but the program design requires states to chip in 10 percent of the costs. As with the pre-k design, this state cost-sharing requirement could cause Republican-led states to not participate in the program.

Unlike the universal pre-k program, which is available to all children regardless of the parent’s economic circumstances, eligibility for child care subsidies is subject to an income test, an asset test, and an activity test.

Activity Test

The activity test requires parents to be engaged in one of the following:

  1. Full-time or part-time employment.
  2. Self-employment.
  3. Job search activities.
  4. Job training.
  5. Education.
  6. Work-limiting health treatment.
  7. Activities to prevent child abuse, neglect, or family violence.
  8. Work or training activities related to SNAP or TANF requirements.
  9. Taking leave under the FMLA or a paid leave program.

This activity test is not applied to kids with a parent over the age of 65. It also does not apply to “vulnerable children identified by the lead agency involved, which at a minimum shall include children experiencing homelessness, children in foster care, children in kinship care, and children who are receiving, or need to receive, child protective services.”

For single-parent families, the one parent needs to satisfy the activity test. For two-parent families, only one of the two parents needs to satisfy the activity test.

This is a completely insane eligibility rule. It reads as if someone initially had the idea to say child care should only be available to working parents, but then lawmakers began realizing there are lots of situations where you would not want to exclude nonworking parents, and so the list became so long that it appears to include basically everyone except parents with work-limiting disabilities, such as those on SSI or SSDI. Of course the main people who will be impacted by the test will be those who actually satisfy it but who struggle to fill out the paperwork proving their status, such as those working in informal sectors.

The CRS took a stab at estimating what percent of parents would be excluded by the activity test. After admitting that it doesn’t really know because there is no dataset that would allow you to compute such a thing, they settle on saying that around 5 percent of kids would be ineligible for child care benefits due to their parents failing the activity test.

In reality, of course, “parents who fail the activity test” will not be a static population. People who are employed one day can be unemployed the next. In that case, the activity test would seem to require that they inform their child care provider of their job loss and promise to them that they are now engaged in job searching or else risk losing their child care benefits. Is that what you really want to be doing with your life after you lose your job? Telling your child care provider that you are being a good boy?

Nothing good will come of this activity test. It’s cruel and pointless.

Asset Test

The asset test excludes children who live in families with more than $1 million of assets. The $1 million figure is so high that the CRS does not even bother to figure out how many people would be excluded by it and instead assumes that everyone who satisfies the income test (discussed below) also satisfies the asset test.

According to the 2019 Survey of Consumer Finances (SCF), around 14.2 percent of all families have $1 million or more in assets. But the vast majority of families with that many assets are beyond the age where they would be using child care services. Among families with heads aged 40 and below, only 3.5 percent have assets that high.

But this does not mean that this asset test excludes 3.5 percent of parents. To be eligible for benefits, you also have to have an income below a certain level. The precise level will vary state to state, but in almost all cases, a family with children young enough to be in child care that has over $1 million in assets is also going to have an income so high that they won’t be eligible for much if any benefits.

In the 2019 SCF, only 1.3 percent of families with heads aged 40 and below have both $1 million of assets and an income below $200,000. Only 0.6 percent with assets that high have an income below $150,000.

This asset test does not actually save any meaningful money. It’s just going to be a hassle for all parents to have to prove that they don’t have $1 million of assets.

Income Test

The income test is implemented as a copayment scheme in which families are required to pay a certain percentage of their income towards the unsubsidized price of child care with the government picking up the rest of the bill.

The income-based copayments are as follows:

Income as Percent of State MedianCopayment as Percent of Income
Less than 75%0%
75% to 100%0% to 2%
100% to 125%2% to 4%
125% to 150%4% to 7%
Over 150%7%

Starting in 2025, every family will be eligible for child care subsidies subject to these income-based copayments. But in 2022, 2023, and 2024, eligibility is restricted to families below a hard income cutoff.

YearEligible Incomes as Percent of State Median Income
20220% to 100%
20230% to 115%
20240% to 130%
2025All incomes

This means that, in year 2022, if you make $1 above your state’s median income, you will go from paying 2% of your income towards child care to paying the entire unsubsidized child care bill. This benefit cliff will be pushed to $1 above 115% of your state’s median income in 2023 and then $1 above 130% of your state’s median income in 2024 until being eliminated in 2025.

Needless to say, having these benefit cliffs during the first few years is not a good design and is certain to cause frustration and resentment among users.

The subsidies for the ACA exchanges were initially designed with this exact kind of benefit cliff for people with incomes $1 above 400 percent of the federal poverty line. This was so bad that the Democrats temporarily eliminated the ACA exchange subsidy cliff in the American Rescue Plan Act and are planning to permanently eliminate it in the very same reconciliation bill where they are introducing these virtually identical benefit cliffs for their child care subsidy scheme.

After the benefit cliffs are eliminated in year 4 of the program, what we are left with is a copayment scheme that essentially functions as a progressive income tax on users of child care services, except that the amount of “child care tax” any particular family pays is capped at the unsubsidized cost of child care.

To see how this plays out in hard numbers, I looked at Texas because it is the median state by median income. In Texas, in 2019, the median cash income was $63,000, according to the CPS ASEC, while the base quality price of toddler care was $9,800, according to the Center for American Progress.

The following graph shows how much each family would have paid for child care in Texas using the copayment rules for the years without the benefit cliffs.

Here is the same graph, but done as a percent of income instead of with dollar figures.

Because the “child care tax” is capped at the unsubsidized price of care, the rates initially rise with income but then, after a certain income level, begin to decline. The graph above goes to $250,000 of income, at which point a family is paying less than 3 percent of their income towards child care. If you continued up the income ladder, the percentage gradually approaches 0%.

The question you have to ask yourself about these copayments is what is the point of including them? Grades K-12 do not have copayments. The Democrats pre-k proposal does not have copayments. So why are they being imposed on ages 0 through 2?

Ostensibly the purpose of the copayments is to “save money.” But it doesn’t really save that much money and clearly using a real tax would be better than using this copayment, which again is essentially an income tax on child care users.

To see what the difference is, I used the 2019 CPS ASEC along with CAP’s estimate of the price for base quality child care for every state, to determine how much it would cost if every single child between the ages of 0 and 2 were in child care and the federal government paid the full cost. According to my calculations, the number is $166 billion, which is around 0.8 percent of GDP. Applying the copayments only brings the cost down to $85 billion, which is 0.4 percent of GDP and even that is misleading because it’s not so much that it brings the “cost” down but rather that it “pays for” some of the cost via a child care tax.

If you assume only half of parents will use the system, which most cost estimates tend to do (newborns don’t use the system and many parents prefer to care for young children in the home), then the numbers in the above paragraph are of course cut in half.

The small difference between having copayments and not having copayments could easily be made up for by using a real tax or by very slightly trimming down the military budget.

Eliminating the copayments would allow us to save hundreds of millions of pages of paperwork each year and millions of headaches caused by people trying to prove their income for the copayment scheme. It would also make the scheme fairer because the proposed copayment rules create a number of bizarre asymmetries (e.g. under the copayment scheme, parents of twins will pay half as much for child care over the course of their lives as parents who have two kids spaced 3 years apart).


As with everything the Democrats put out, one is left puzzled by many of the details of the plans. Extending K-12 down to age 0 is a very good idea and one that is on its face very simple to do and explain. Yet Democrats just can’t seem to resist taking that idea and screwing it up.