In early September—in the midst of a family separation crisis of its own making, and with hundreds of children still separated from their parents1—the Trump administration published a notice of proposed rulemaking2 to dissolve the two-decades-old Flores settlement, which sets standards for the treatment of children in immigrant detention.3 At its most basic, the proposed rule would provide the Trump administration with a way to get around the requirement that children not be kept in secure, unlicensed detention facilities for more than 20 days.4 This requirement, and the Flores settlement in general, has, to date, prevented the Trump administration from indefinitely incarcerating children alongside their parents in unsafe, inappropriate conditions.
The Department of Homeland Security (DHS) has previously tried and failed to get its family incarceration facilities—known as “family residential centers” (FRCs)—licensed by an appropriate state agency, as required by the settlement.5 Likewise, DHS has tried and failed to get the judge overseeing the Flores settlement to let it out of the licensing requirement with respect to children detained with a parent, which would make it possible for the department to detain accompanied children indefinitely.6 Now, rather than continue to pursue state licensing, the Trump administration has proposed a rule that would allow DHS itself to pursue “alternative licensing.”7 Under the proposed rule, the department would be authorized to write its own standards and identify its own external auditors, who would then be charged with certifying that these facilities can reasonably hold and protect children. Given that, as recently as June 2018, the DHS inspector general found that “ICE’s inspections and monitoring of detention facilities do not lead to sustained compliance or systemic improvements,”8 and given the widespread abuses already occurring in family detention,9 DHS should not be granted new authority to certify conditions for the protection of children in its own facilities.
In the proposed rule, the administration is clear that the DHS “alternative licensing process” may extend the detention of some children—along with their parents—in family incarceration facilities.10 However, the proposed rule fails to mention the ample research that demonstrates that incarcerating children for any length of time is detrimental to their health and well-being.11 Put simply, children should not be locked up.12 The Trump administration has at its disposal a range of alternatives to detention that are far more humane and cost a fraction of what it takes to incarcerate someone while still ensuring that families complete the legal process to make their case for protection.13 With this proposed rule, DHS is choosing to pursue maximum detention at maximum cost—and with minimal oversight.
In publishing its proposed rule, the Trump administration declined to estimate the potential costs, even though this is required under a long-standing executive order.14 For example, while DHS argues that children and parents would likely be held for longer under the proposed rule, it states that “ICE is unable to estimate how long detention would be extended for some categories of minors and their accompanying adults.”15 In essence, the proposed rule argues that historical data on the number of people kept in custody, their average lengths of stay, and more are subject to too many external factors to use for future projections. DHS also does not project the number of additional family facilities it might build once freed from the requirement to seek state licensing. Yet the logic of the proposed rule implies that, in order to hold more family members, the department would need to obtain more bed space than the current 3,326-person capacity of its three existing facilities.16
In the proposed rule, DHS presents data from as recently as fiscal years 2014 and 2015, before a federal court affirmed that the Flores protections apply to children held in family detention. These data provide a clear window into what a world without Flores might look like. Given the available data, the administration’s excuses for not attempting a cost estimate of the proposed rule appear weak at best.17
This issue brief runs the numbers on what the implementation of the proposed rule would likely cost DHS. It first calculates DHS’ additional detention bed space needs once the department is freed from the current 20-day limitation, putting a dollar figure on just how expensive these additional beds would be. The brief then estimates how many new family residential centers the department would need to acquire in order to hold these additional children and families, before calculating the start-up costs of these new facilities. Finally, while the government has not released enough data to calculate the proposed rule’s potential costs to the Department of Health and Human Services (HHS)—particularly the Office of Refugee Resettlement (ORR), which is charged with the care and custody of unaccompanied children—the brief details some of the ways in which HHS could also incur high costs.
Proposed rule could cost DHS as much as $12.9 billion
Using the government’s own data, the Center for American Progress estimates that, over a decade, the proposed rule would cost DHS slightly more than $2 billion at the low end, and as much as $12.9 billion at the high end.18 On an annualized basis, the low-end estimate would cost $201 million each year, and the high-end estimate would cost slightly less than $1.3 billion each year.
These costs include two parts: first, annual costs from the additional detention beds that would be necessary under the rule as more children and parents are detained together for longer; second, one-time start-up costs for DHS to acquire additional family residential centers in order to incarcerate these children and parents for longer. These calculations do not take into account any costs to HHS under the rule, which could easily run into the millions of dollars.
Even under the most conservative estimate, the proposed rule would far exceed the $100 million threshold by which a regulation would be considered economically significant, thus triggering additional review and analysis—which DHS and HHS declined to undertake.19 These high costs would also render the proposal a “major rule” under the terms of the Congressional Review Act—a designation that would trigger both a Government Accountability Office (GAO) report prior to the rule’s implementation and additional time before the rule takes effect, giving Congress the opportunity to take action to stop it.20
The proposed Flores rule is unlikely to deter future family members from arriving
In justifying the need for the proposed rule—and with it, the indefinite incarceration of children alongside their parents—the Trump administration has argued that its inability to detain families for longer than 20 days in secure, unlicensed facilities has led to many families being released, thus creating “a powerful incentive for adults to bring juveniles on the dangerous journey to the United States.”21 However, as University of California, San Diego professor and CAP senior fellow Tom K. Wong has illustrated, the evidence does not support the idea that overturning Flores will serve as a deterrent to future migration. In his analysis of family detention data, Wong found, “The expanded use of family detention is not statistically significantly related to decreases in the monthly number of U.S. Border Patrol apprehensions of families at the southwest border.”22 Likewise, Wong’s research finds no statistically significant increase in apprehensions of families at the border after July 2015. This finding disproves DHS’ claims in the proposed rule that the court ruling—which confirmed that the terms of the Flores agreement applied to accompanied children in family incarceration—has led to more families coming to the United States.23
In addition, it is unlikely that overturning the Flores settlement will lead to fewer children and families seeking asylum in the United States, given the harrowing conditions in countries like El Salvador and Honduras that are pushing people to seek protection here.24 For these reasons, the analysis in this brief assumes that future migration patterns will continue apace.25
Additional bed costs under the Flores rule
Under the proposed rule, there are at least two ways that U.S. Immigration and Customs Enforcement (ICE) could incur higher costs relating to the detention of children and families in family residential centers:
- Costs that stem from detaining people for longer than they are currently being held
- Costs that stem from detaining more people than are currently being held
In estimating additional bed costs under the Flores rule, the Center for American Progress considered two scenarios.
In Scenario 1, the overall number of children and families stays largely the same as it was in FY 2017—the most recent year for which data are available.26 However, the length of their stay in detention is increased to the historic FY 2014 level. (see the Appendix for more information on the methodology)
Scenario 1 includes four primary assumptions:
- The number of people booked into FRCs remains the same as in FY 2017, during which 37,825 individuals were incarcerated in family facilities. This assumption is likely conservative, given that significantly more families arrived in FY 2018 than in FY 2017, meaning that the baseline FRC population for FY 2018 is likely higher too.27
- The length of stay under the rule will revert back to the historical average. In FY 2014, individuals were held in FRCs for 47.4 days, on average.28 This changed after a 2015 ruling from Judge Dolly M. Gee, the district court judge overseeing the Flores settlement, which held that children cannot be held in unlicensed secure detention facilities for more than 20 days.29 As of FY 2017, the average length of stay had dropped to only 14.2 days.30 By dissolving the Flores settlement and allowing the federal government to self-certify facilities, this analysis assumes that the average length of stay will revert, at a minimum, to FY 2014 levels.31
- DHS estimates that, had the proposed rule been in effect in FY 2017, an additional 2,787 children—including children who received negative credible fear determinations or final orders of removal—would have been held in detention for longer. Although, in the proposed rule, DHS explicitly declines to estimate how much longer these children would be held in FRCs, this analysis assumes that their average lengths of stay would increase to 25 days.32
- The daily cost per family detention bed, which ICE estimated in its FY 2019 congressional budget justification, remains the same: $318.79.33
Importantly, ICE’s three family residential centers—South Texas Family Residential Center (Dilley), Karnes County Residential Center, and Berks County Residential Center—operate on fixed-cost contracts.34 This means that the price per bed is the same regardless of whether these facilities are half full or at their total capacity of 3,326 beds.35 Therefore, to ensure the most conservative estimates, these calculations assume that ICE begins to incur new costs only after it reaches maximum capacity—in other words, only for the 3,327th bed and beyond.
Under Scenario 1, the annual additional detention costs under the rule would be slightly more than $194 million. (see Appendix for more details)
That being said, the proposed rule would not simply allow the administration to hold children and families for longer; it would also allow them to detain a much greater number of families. Members of the Trump administration, such as Attorney General Jeff Sessions and Secretary of Homeland Security Kirstjen Nielsen, have been clear that their goal is to apply the maximum consequences to the maximum number of people.36 However, a key restraint in their ability to do so has been the 20-day limitation for holding children in secure, unlicensed facilities. Without this, there is every reason to believe the administration will pursue widespread incarcerations.
For this reason, Scenario 2 assumes that every person arriving in a family unit is incarcerated and that their length of stay is increased to the historical FY 2014 level. Under this scenario, ICE would need significantly more detention beds than it currently has; in FY 2018, 107,063 people were apprehended in family units.37 As such, the annual additional detention costs under the rule would be as high as $1.24 billion. (see Appendix)
It is important to note that both of these scenarios envision that the government reverts back to FY 2014 lengths of stay. However, if the Trump administration even increases the time that families stay in detention past the 47.4-day mark, the costs enumerated above would only increase.
The cost of new family residential centers
Because both of the above scenarios go above the current 3,326-family-detention-bed capacity, ICE would inevitably have to either build or acquire new family residential centers.38 However, in the proposed rule, DHS abdicates its responsibility to estimate the cost of new facilities, simply stating that it “is unable to determine how the number of FRCs may change due to this proposed rule.”39
To estimate the cost to ICE of acquiring new bed space, this analysis uses the number of additional detention beds identified in the two scenarios above and calculates the costs based on the two most reasonable routes that ICE can take: one, acquiring new facilities that look more like the Karnes County Residential Center—which has 830 beds—or two, acquiring facilities that look more like the Dilley, which has 2,400 beds.40 The conversion of the Karnes facility to an FRC cost $36 million,41 and the opening of the Dilley facility cost $104 million.42
At the lowest end, acquiring the new facilities needed to detain additional families—as estimated above—would require one-time start-up costs of at least $72 million and as much as $520 million at the highest end.
Under Scenario 1—where the number of people in family incarceration largely stays the same as in FY 2017, but the lengths of stay increase—ICE would need to acquire 1,669 new family detention beds annually, on top of its current maximum capacity.43
- If ICE chooses to build facilities the size of Karnes in order to house these additional detention beds, it would require two new FRCs, at a total start-up cost of $72 million.
- If ICE chooses to build facilities the size of Dilley, it would require one new FRC, at a start-up cost of $104 million.44
Under Scenario 2—where ICE detains every person in a family unit and detains them for longer—ICE would need to acquire 10,660 new family detention beds annually, on top of its current maximum capacity.45
- If ICE chooses to build facilities the size of Karnes, it would require 13 new FRCs, at a total start-up cost of $468 million.
- If ICE instead chooses to build facilities the size of Dilley, it would require five new FRCs, at a total start-up cost of $520 million.46
Potential additional costs to HHS
The calculations above only estimate detention costs to ICE to implement the proposed rule. In addition, however, HHS—and particularly the ORR, which runs the Unaccompanied Alien Children program—would likely incur new costs under the proposed rule. While HHS has not provided sufficient data to estimate a dollar figure for these additional costs, it is important to note that, if the rule goes into effect, these costs would nonetheless be incurred.
Over the past year, the ORR’s costs have been rising sharply. As The New York Times and CNN recently reported, in the past few months, the number of children in ORR care has skyrocketed: from only 2,400 in May 2017 to nearly 13,000 children in custody as of September 2018. Meanwhile, the average length of stay in ORR care has also risen. As of September, the average length of stay was 59 days, compared with 48 days in FY 2017.47
These increases are not due to a new influx of unaccompanied children entering the country. Rather, under the Trump administration, the ORR is releasing far fewer unaccompanied children to sponsors, and an agreement between ICE and HHS to share information on potential sponsors is having a chilling effect on these sponsors coming forward.48 Just between July and late September, ICE arrested 41 potential sponsors.49 Rather than have sponsors assume the burden of caring for these children as they await their day in court to make their claim for asylum or other forms of relief from deportation, the government is keeping these children in custody at high costs.
To deal with the rising numbers of children being held in custody, the Trump administration recently announced that it would triple the size of an emergency tent city in Tornillo, Texas.50 However, at $750 per bed, per day, the emergency facility’s beds cost nearly three times more than nonemergency ORR shelter beds.51 To pay for these additional costs, in early September, HHS Secretary Alex Azar sent a letter to Congress stating his intention to take $266 million away from programs such as the National Cancer Institute and the Ryan White HIV/AIDS program.52
Beyond the higher costs that HHS is already bearing, there is reason to believe that the proposed rule would impose even greater burdens. In particular, the rule makes a series of changes to when and how unaccompanied alien children (UACs) can be held in secure facilities, which function more like juvenile jails than shelters.
In section 410.203 of the proposed rule, HHS sets out the criteria for placing children in secure facilities. It expands the population of children who might be placed in these facilities to include any child who “engages in unacceptably disruptive behavior that interferes with the normal functioning” of the shelter, as well as any child that “displays sexual predatory behavior.”53 A separate section of the proposed rule—Proposed 8 CFR 236.3(n)—gives HHS new authority to “reassum[e] custody of previously released minors if they become an escape-risk, become a danger to the community, or are issued a final order of removal after being released.”54
Based on information HHS has made publicly available, it is impossible to know how many more children would be placed in secure facilities because of these sections of the proposed rule. However, placing children in secure facilities could lead to large costs. For example, secure bed space in the Yolo County Juvenile Detention Center in California costs $651 per bed, per day—more than 2.5 times the cost of the average ORR shelter bed.55 Not only are these beds more expensive, but on average, children spend nearly four times longer in secure facilities.56 In total, given the higher per-bed costs and much longer lengths of stay, the cost to hold one child in secure care is nearly 10 times the cost to hold a child in regular shelter care.57
HHS argues that it “does not expect this proposed rule to impose any additional costs” related to increased detention in ORR facilities.58 However, given the likelihood that costs will, in fact, rise for the ORR, HHS needs to make a full accounting of what the proposed rule will likely cost.
Altogether, at the lowest end, the proposed rule to dissolve the Flores settlement would cost DHS slightly more than $2 billion over a decade. This would include $194 million annually in additional detention bed space and start-up costs of at least $72 million. Meanwhile, at the highest end, this rule would cost DHS an astronomical $12.9 billion over a decade—including $1.24 billion annually in additional detention bed space and a one-time start-up cost of $520 million. Given the cost to the government, as well as the incalculable harm of this mass incarceration to children and families, the Trump administration must not move forward with this rule.
Philip E. Wolgin is the managing director for Immigration Policy at the Center for American Progress.
The author thanks Sam Berger, Michelle Brané, Diane Eikenberry, Sharita Gruberg, Tom Jawetz, Scott Nathan, Katharina Obser, Ursela Ojeda, Ben Olinsky, and Nicole Prchal Svajlenka for their help with these calculations.
This appendix explains how the cost estimates above were calculated.59
Definitions and acronyms
- Average daily population (ADP): the average number of people in detention on any given day
- Average length of stay (ALOS): the average number of days that an individual stays in a detention facility, from the time they are booked in to the time they are released
- Family residential center (FRC): ICE’s name for family incarceration facilities
- Unaccompanied alien child (UAC): the term for an individual under the age of 18 who lacks lawful immigration status and arrives, or is apprehended, in the United States without a parent or legal guardian60
Scenario 1: Current detainees are held for longer
The formula for calculating the costs for added detention is as follows:
Scenario 1 Cost = (New ADP – FRC Current Capacity) × Cost Per Detention Bed × 365 Days
Scenario 1 Cost = $194,100,000 = (4,995 – 3,326) × $318.79 × 365
Breaking this down, calculating New ADP is the sum of three parts:
New ADP = 4,995 = FY 2017 ADP + Total Additional ADP A + Total Additional ADP B
In this case, the FY 2017 ADP of 1,472 is extrapolated from the fact that 37,825 family members were booked into FRCs in FY 2017, with an average length of stay of 14.2 days.
FY 2017 ADP = 1,472 = (37,825 × 14.2) / 365 Days
To get to Additional ADP A, we assume that each of the 37,825 people booked into an FRC in FY 2017 would have their stay lengthened from 14.2 days to the FY 2014 historical average of 47.4 days.
Additional ADP A = 3,441 = (37,825 × (47.4 – 14.2)) / 365 Days
To get to Additional ADP B, we take the four groups of children who DHS estimates would have been held for longer in FY 2017 had the rule been in place—a total of 2,787 children—and estimate that their average length of stay would be increased from 14.2 days to a total of 25 days.61
Additional ADP B = 82 = (2,787 × (25 – 14.2)) / 365 Days
Scenario 2: Every person in a family unit is incarcerated
As with Scenario 1, the formula for costs in this scenario is as follows:
Scenario 2 Cost = (New ADP – FRC Current Capacity) × Cost Per Detention Bed × 365 Days
Scenario 2 Cost = $1,240,400,000 = (13,986 – 3,326) × $318.79 × 365
Here, the New ADP is much higher since the assumption is that all children and families who arrive together are put into family detention, at longer lengths of stays.
New ADP = 13,986 = FY 2018 ADP + Total Additional ADP B
In this scenario, the FY 2018 ADP is calculated as follows: In FY 2018, U.S. Customs and Border Protection apprehended 107,063 people in family units.62 This number is then multiplied by the FY 2014 ALOS of 47.4 days as such:
FY 2018 ADP = 13,904 = (107,063 × 47.4) / 365 Days
Additional ADP B, of 82, is calculated as in Scenario 1 above.
10-year and annualized costs
Calculating the total low- and high-end cost to DHS from the proposed rule is as follows:
10-Year Costs = (Cost of Additional Detention Beds × 10) + Cost of Acquiring New FRCs
Low-end: $2,013,500,000 = ($194,100,000 × 10) + $72,000,000
High-end: $12,923,800,000 = ($1,240,400,000 × 10) + $520,000,000
Annualized Costs = 10-Year Costs / 10
Low-end: $201,300,000 = $2,013,500,000 / 10
High-end: $1,292,400,000 = $12,923,800,000 / 10