Reforming Childcare Allocation Through Centralized Matching

---

By: Tyler Kim

Edited By: Stephen Shiwei Wang

---

2023 was a historic year for South Korea, as it recorded the world’s lowest fertility rate, with a record TFR (Total Fertility Rate) of 0.72 Children per Woman, reflecting the largest demographic in the nation’s history.¹ Despite this crisis, families navigated a childcare shortage where the average wait time for daycare was 9.2 months, leaving 33.5 percent of infants on waitlists.² Uncertainty in accessing childcare raises the expected cost of childbearing, prompting many families to delay or forgo having children. This outcome is driven by the i-Sarang portal, which is an established Point-Based Priority System acting as an inefficient queue with no linkage between family needs and center capacity. It’s a “Coordination Trap” in which the fear of no available childcare pushes parents (mostly mothers) into unwanted career breaks, ultimately deterring them from having children.² I argue for a centralized matching system as a solution to this dilemma with the use of  Gale-Shapley Deferred Acceptance algorithm.³ By applying this algorithmic market design concept, the current national “childcare queue” issue could be addressed to efficiently match families to centers based on more informed preferences in a strategy-proof manner.⁴

Flaws in South Korea’s Centralized Daycare Allocation Mechanism

The South Korean government originally intended national daycare policies to create a national computer-based service (i-Sarang portal), replacing the previously localized admissions system and standardizing the allocation process.⁵ The Point-Based Priority System operates as digital portals for childcare centers (Eurinijip) and Cheoeum-Hakeyo for kindergartens (Yuchiwon), which are intended to allocate access to limited daycare spots in a fair manner, using factors related to socioeconomic needs rather than by price or lottery system.⁶

Priority (First Class) selection is currently given to families possessing the most opportunity cost for their care, which is assigned to different households through points based on double-income, single-parent, and multiple children, etc.⁷ By opportunity cost, it refers to families who would otherwise bear an above-average cost for childcare if public childcare is inaccessible to them. In theory, this transparent Management Information System (MIS) was designed to tie government subsidies to provider capacity, enabling female labor participation and facilitating social protection.^{8, 9}

The priority selection mechanism is an attempt at needs-based childcare resource allocation, but the entire administration problem is far from over. To secure childcare, parents are forced to register with numerous providers,  creating a “Phantom Demand” that obscures true market needs. Consequently, while public daycare centers (Gong-rip) face years-long waiting lists, neighboring privately owned places (Min-gan) may have vacancies, highlighting a lack of coordinated supply and demand across the market.

Due to the randomness of the selection process, mothers often face involuntary career breaks. This created a serious “Motherhood Penalty”, which has been identified as one of the leading causes of demographic collapse. To be precise, the “Child Penalty”, which includes job and wage loss, is responsible for about 40% of the reduction in fertility.¹⁰

Market Failure and Structural Gaps

The current system fails on efficiency by misallocating capacity and on equity by penalizing families who report preferences truthfully. Due to South Korea’s flawed and nationalized daycare system, the public childcare sector has shifted from a market governed by supply and demand to a matching market defined by government-set prices.^{11, 12} With prices unable to rise to clear the market, the system must rely on a rigid allocation mechanism. The current system has three critical structural gaps: a lack of strategy-proofness, a high Price of Anarchy (PoA), and a “Coordination Trap”.¹⁴ The most direct structural gap is the system’s non-strategy-proofness. By clearing the market piecemeal for each center in isolation, rather than for the market as a whole, the portal functions as a decentralized, center-level priority queues.¹³

The non-strategy-proof market penalizes truthful reporting by families: ranking a competitive public center first risks losing a seat at a viable second choice to a strategic reporter.^{15, 16} Anticipating this, rational families have incentives to “game” the system by “stockpiling” applications on multiple waiting lists as an insurance policy against uncertainty. Defensive stockpiling behavior by families leads to a colossal “phantom demand”; this congestion masks the true distribution of need and prevents efficient market clearing.¹⁷

Consequently, this created a high Price of Anarchy, whereby misalignment between public (gong-rip) and private (min-gan) capacity has forced the daycare market into a “Coordination Trap”. This difficulty in access acts as a negative signal, involuntarily forcing mothers out of the labor market and driving the “Child Penalty”.¹⁸

Policy Proposal

To resolve the “Waitlist Paradox” and any inefficiencies in the i-Sarang queue, this paper calls for a revenue-neutral structural reform: A move from South Korea’s existing decentralized mechanism to a Centralized Matching Clearinghouse using the Gale-Shapley Deferred Acceptance (DA) algorithm.¹⁹ Rather than applying to individual centers separately, parents submit a single Ranked Order List (ROL) of preferred providers. The deferred acceptance algorithm then evaluates parents’ preferences simultaneously against existing priority points, such as dual-income status, in a sequence of rounds.

It can be argued that the key economic advantage of this shift is the Strategy-Proofness of the DA mechanism, which renders truth-telling to be a dominant strategy for parents. By listing a competitive public center as their first choice, families do not jeopardize their ability to match with a “safe” second choice under the DA mechanism, compared to the status quo Mechanism.

This creates a strong incentive to reduce seat hoarding and massive phantom demand that masks true need. By facilitating a Stable Match in which no family has justified envy for another, the government could optimize the existing capacity in private and public systems.^{20, 21} This approach addresses the time poverty driving the motherhood penalty and removes barriers to national fertility.²²

How the Matching System Would Work

1. Parents provide a single, Ranked Order List (ROL) to the i-Sarang website, listing their ranked preferences of centers and confirming their priority, in one secure transaction.
2. Inputs from providers and families are processed through a central government clearinghouse with the Gale-Shapley Deferred Acceptance (DA) algorithm, creating an immediate, stable, strategy-proof match for the entire market.
3. The current digital interface is updated to allow families to submit a single Ranked Order List immediately upon birth registration tying demographic data to the matching engine.
4. A brief “Resolution Window” is provided allowing families the chance to contest their placement in the case of extreme, unforeseen circumstances (verified family relocation or medical need, for example).
5. To offset overcrowding in public centers, top-performing private centers are provided with central government public-track funds, either for meeting quality benchmarks or accepting high-priority infants.²³ Tests in other countries and research in academic journals illustrate that matching systems increased their respective national attributes.^{24, 25} A matching system would substantially reduce effective waiting times.^{26, 27} Providers would equally benefit from better matching: planning with more certainty, and new digital infrastructure would free them from long waitlists and provide higher-quality care.²⁸

Policy Recommendations

• Deploy in multiple markets- including non-capital regions like Gangnam-gu, Sejong City, and Jeonbuk- to stress-test the algorithm against varied regional supply-demand imbalances and demographic demands.
• Set up a team within the Ministry of Health and Welfare to own the algorithm, audit capacity numbers, and maintain privacy parameters.

• Revise the portal to enable immediate ROL submission with birth registration, bypassing the pass/fail registration step and feeding directly into the matching engine
• Link public-track central government subsidies to (Min-gan) private centers that prioritize enrolling children from the public waitlist; essentially this commercializes the acceleration of the public queue.
• Annually publish a “Hot-Spotting” analysis based on centralized data to pinpoint specific childcare deserts, enabling targeted physical supply expansion, beyond just algorithmic modeling.

South Korea’s childcare crisis is a coordination failure, not just a capacity shortage. The current allocation system distorts demand, increases uncertainty and reinforces labor market exit among mothers. A centralized, strategy-proof matching system would correct these failures without increasing public spending, mitigate the motherhood penalty, and remove a structural barrier to fertility. Childcare allocation reform is, therefore, not only an administrative upgrade but also a demographic and labor market intervention.

---

 

Work Cited

1. OECD. 2024. Economic Survey of Korea 2024. OECD Publishing.
2.  Korea Development Institute. 2023. Impact of childcare subsidies on fertility. KDI
3. Gale, D., & Shapley, L. S. 1962. College admissions and the stability of marriage. The American Mathematical Monthly, 69(1), 9–15
4. Abdulkadiroğlu, A., & Andersson, T. 2003. 2023. Chapter 3 – School Choice. Handbook of the Economics of Education. Elsevier, volume 6. Page 135 – 185. https://www.sciencedirect.com/science/article/pii/S1574069222000010.  
5. OECD. 2020. Strengthening early childhood education and care in Korea. OECD Publishing.
6. UN Women. 2022. Progress of the world’s women: Care in a changing world. United Nations.
7. Korea Development Institute. 2023. Impact of childcare subsidies on fertility. KDI
8. World Bank. 2024. Female labor force participation in the Republic of Korea: Economic and human capital utilization. World Bank
9. Roth, A. E. 2008. What have we learned from market design? The Economic Journal, 118(527), 285–310.
10.  Korea Development Institute. 2023. Impact of childcare subsidies on fertility. KDI
11. Roth, A. E. 2008. What have we learned from market design? The Economic Journal, 118(527), 285–310.
12. The Royal Swedish Academy of Sciences. 2012. Stable matching and market design:Scientific background on the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel. Nobel Prize Outreach.
13. Roth, A. E. 2008. What have we learned from market design? The Economic Journal, 118(527), 285–310.
14. OECD. 2020. Strengthening early childhood education and care in Korea. OECD Publishing.
15. Abdulkadiroğlu, A., & Andersson, T. 2003. 2023. Chapter 3 – School Choice. Handbook of the Economics of Education. Elsevier, volume 6. Page 135 – 185. https://www.sciencedirect.com/science/article/pii/S1574069222000010. 
16. The Royal Swedish Academy of Sciences. 2012. Stable matching and market design:Scientific background on the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel. Nobel Prize Outreach.
17. Pathak, P. A. 2011. The mechanism design approach to student assignment.
18. OECD. 2024. Economic survey of Korea 2024. OECD Publishing.
19. Gale, D., & Shapley, L. S. 1962. College admissions and the stability of marriage. The American Mathematical Monthly, 69(1), 9–15
20. OECD. 2020. Strengthening early childhood education and care in Korea. OECD Publishing.
21. Roth, A. E. 2008. What have we learned from market design? The Economic Journal, 118(527), 285–310.
22. UN Women. 2022. Progress of the world’s women: Care in a changing world. United Nations.
23. OECD. 2024. Economic survey of Korea 2024. OECD Publishing.
24. Pathak, P. A. 2011. The mechanism design approach to student assignment.
25. Correa, J., Epstein, R., & Yumbla, G. 2019. The school admission system in Chile: Theoretical and practical challenges. Research Paper.
26.  Korea Development Institute. 2023. Impact of childcare subsidies on fertility. KDI
27. World Bank. 2021. Digitalizing social protection: A framework for building management information systems. World Bank.
28. International Labour Organization. 2018. Care work and care jobs for the future of decent work. ILO.

---