Universal Basic Capital
with a Social Wealth Fund

Very Preliminary

Sang Yoon (Tim) Lee

QMUL, TSE and CEPR

Alexandre Gaillard

Brown University

August 13, 2026

Motivation

Motivation

• Technology expected to widen inequality in both labor income and asset returns
• Renewed interest in UBI as a response
  • Existing literature is negative on effectiveness: too costly and distortionary
  • even more so with forward-looking education and entrepreneurship decisions
• Can Universal Basic Capital (UBC)
  financed by a Social Wealth Fund (SWF)
  be a more efficient mechanism for redistribution?

Real Motivation

© Getty Images

Mechanism and Preliminary Results

1. UBC: pledge future UBI at “early in life” decision nodes

2. SWF: returns from publicly owned business capital in lieu of distortionary taxation

   • Real-world analogues
   • But public capital crowds out private capital

3. Quantitative question: can these mechanisms overturn UBI’s negative effects?

   • Effects on output and physical and human capital
   • Incidence across workers and entrepreneurs
   • Interaction between the two mechanisms

UBC: Pledgeability of Future Transfers

Let \(\Omega_s:=\) present value of future transfers at time \(s\), and define \[ a_{s+1}\ge-\frac{\color{red}{\phi_h}\Omega_{s+1}}{1+r}, \qquad k\le\lambda(a+\color{red}{\phi_k}\Omega_s). \]

• \(\phi_h>0\): relax borrowing constraint, can use for education
• \(\phi_k>0\): relax collateral constraint, can use for production
• Pledgeability relaxes the constraint before education and production choices

SWF: Less Distortionary but Crowds Out Capital

• SWF owns share \(\kappa\) of entrepreneurial capital: \[ (1-\color{red}{\kappa})k\le\lambda(a+\phi_k\Omega_s). \]
• Receipts can finance universal dividends or capital transfers
• Pools entrepreneurial risk, replaces distortionary taxation

Tradeoff: lowers private capital financing, leading to ambiguous GE effects

Four Policies, Two Margins

Policy	Pledgeability?	SWF?
UBI	No	No
UBC	Yes	No
SWD	No	Yes
SWC	Yes	Yes

Simple Model

Minimal Environment

• OLG: one dynasty = adult + child
• Adult state: assets \(a\), human capital \(h\), entrepreneurial ability \(z\)
• Occupation: worker \(W\) or entrepreneur \(E\)

\[ V(a,h,z)=\max\{V^W(a,h,z),\mathbb EV^E(a,h,\epsilon z)\}, \] - Child investment \(e\): \[ h'\sim H(\cdot\mid h,e),\qquad z'\sim F(\cdot\mid h'). \]

• Key friction: financing at early-life decision nodes

Household problem · Production · Government budget

UBI vs UBC: \(\phi\uparrow\)

Proposition (UBI\(\prec\)UBC) \[ \begin{aligned} V^{UBC}(a,h,z) &\ge V^{UBI}(a,h,z) \\ e^{UBC}(a,h,\tilde z) &\ge e^{UBI}(a,h,\tilde z) \end{aligned} \] with strict gains when early constraints bind on a non-degenerate set.

Corollary (UBI\(\prec\)UBC and bequests) The welfare gap \[ \mathbb{E}\left[ V^{UBC}(a',h',z')-V^{UBI}(a',h',z')\right] \] is larger whenever the bequest constraint binds.

How UBC Gains (Might) Arise

UBI vs SWD: \(\kappa=0\;\rightarrow\kappa>0\)

Lemma:

If \(r>0\), SWF net proceeds imply: \(\omega^S>\omega^U\) since \(k(a,\epsilon z)=0\) if \(R(\epsilon z;w)\leq\bar R\).

Proposition (UBI\(\prec\)SWD) \[ V^{SWD}(a,h,z)>V^{UBI}(a,h,z) \] since larger \(\omega^S\) frees resources.

As in UBI\(\prec\)UBC, welfare gaps are larger if the bequest constraint binds.

Inequality Amplifies the Mechanism

• More dispersion \(\rightarrow\) more households in high-return states, early constraints bind
• High entrepreneurial returns \(\rightarrow\) higher value of collateral
• Human-capital complementarity \(\rightarrow\) higher return to education investments

Quantitative Model

Lifecycle Structure

• All households save at end of period
• All retirees earn pensions; can run business but can’t work
• Timing preserves intuition from simple model GE channels

Calibration

• Fixed and calibrated parameters
• Empirical targets:
  period earnings, entrepreneurship, wealth distribution, education investments
• Benchmark “UBI”: \(4\%\) of annual worker earnings
  • \(13\%\) pledgeable for education and entrepreneurial finance
• Annual safe rate: \(3\%\)
• Government spending: \(18.9\%\) of GDP

Exact calibration in progress

Counterfactual Design

• Holding structural and all other fiscal parameters constant, vary
  1. SWF ownership of business capital (benchmark counterfactual: \(\kappa=0.1\))
  2. transfer pledgeability (\(\phi_h,\phi_k\): benchmark \(\rightarrow 1\))
• Compare:
  1. SWF (\(\kappa\)) proceeds finance UBC (\(\omega\))
  2. SWF (\(\kappa\)) balanced by income tax
  3. UBC (\(\omega\)) financed by income tax
• Solve for \((r,w)\) in stationary GE
• Compare output, income, physical and human capital

Limitations: no CEV welfare or transitions yet

Preliminary Results

Output

SWF-funded UBC

Tax-funded UBC · SWF-funded tax cuts

• SWF beats matched tax finance in all cases
• Under SWF-funded UBC, \(\phi_k\) raises output about \(1\%\) above benchmark
• SWF-funded tax cuts raise output by \(1.17\%\)

Private Capital

• SWF adds \(1\)–\(2\) pp to the decline versus matched tax finance
• \(\phi_k\) deepens crowd-out; SWF-funded tax cuts limit the decline to \(1.43\%\)

Human Capital

• \(\phi_h\) enables education investment; \(\phi_k\) raises demand for human capital
• Human capital rises only when both margins operate

Entrepreneur Shares

• \(\phi_k\) reverses the decline in entrepreneurship
• Both margins: entry \(+0.18\) pp with SWF finance; \(+0.28\) pp with tax finance

Worker Hours

• UBC lowers worker hours by \(1.5\)–\(2.4\%\)
• Pledgeability softens the decline; SWF-funded tax cuts leave hours unchanged

Who Gains?

• \(\phi_k\) turns conditional-income losses into gains for both groups
• Both margins: worker income \(+0.90\%\); entrepreneur profit \(+0.94\%\)

Who Gains under Alternative Financing?

• Tax-funded UBC lowers conditional worker income in all cases
• With \(\phi_k\), conditional entrepreneur profit rises by \(5\)–\(6\%\)
• SWF tax cuts: worker income \(+0.92\%\); entrepreneur profit \(-3.84\%\)

Interactive Effects

• \(\phi_k\) raises output and entrepreneurship despite lower private capital
• \(\phi_h\) mainly raises human capital, especially with \(\phi_k\)
• SWF raises output but dampens entrepreneurship versus tax finance

1. SWF tax cuts raise worker income but lower entrepreneur profit
2. Tax-funded UBC with \(\phi_k\) does the opposite
3. Optimal mix requires welfare analysis

Conclusion

Conclusion

• UBC relaxes early constraints
• SWF substitutes public for private finance
• Preliminary: with SWF funding, \(\phi_k\) reverses the output loss
  • Financing: SWF beats tax finance on output
  • Incidence: SWF tax cuts: workers gain, entrepreneurs lose
  • Interaction: \(\phi_k\) raises activity; \(\phi_h\) raises human capital

Next: CEV welfare, incidence by wealth and earnings, transitions, optimal policy

References

Awal, Farouk, and Aamir Hashmi. 2025. “Universal Basic Income in an Entrepreneurial Economy.”

Chetty, Raj, Nathaniel Hendren, Patrick Kline, and Emmanuel Saez. 2014. “Where Is the Land of Opportunity? The Geography of Intergenerational Mobility in the United States.” Quarterly Journal of Economics 129 (4): 1553–623.

Conesa, Juan Carlos, Wenli Li, and Yikai Li. 2023. “A Quantitative Evaluation of Universal Basic Income.” Journal of Public Economics 219: 104829.

Corneo, Giacomo. 2022. “Progressive Sovereign Wealth Funds.” Journal of Government and Economics 5: 100033. https://doi.org/10.1016/j.jge.2022.100033.

Daruich, Diego, and Raquel Fernández. 2024. “Universal Basic Income: A Dynamic Assessment.” American Economic Review 114 (1): 38–88.

Ghatak, Maitreesh, and Francois Maniquet. 2019. “Some Theoretical Aspects of a Universal Basic Income Proposal.” Annual Review of Economics 11: 895–923.

Heathcote, Jonathan, Kjetil Storesletten, and Giovanni L. Violante. 2017. “Optimal Tax Progressivity: An Analytical Framework.” Quarterly Journal of Economics 132 (4): 1693–1754.

James, Alexander, and Raymond MacDermott. 2022. “Sovereign Wealth Funds in Theory and Practice.” Annual Review of Resource Economics 14: 447–70.

Jones, Damon, and Ioana Marinescu. 2018. “The Labor Market Impacts of Universal and Permanent Cash Transfers.” American Economic Journal: Economic Policy 10 (4): 1–28.

Luduvice, Lucas. 2024. “The Macroeconomic Effects of Universal Basic Income Programs.” Journal of Monetary Economics 137: 1–23.

Megginson, William, and Li You. 2014. “State-Owned Enterprises and Privatization.” Foundations and Trends in Finance 9 (1–2): 1–153.

National Center for Health Statistics. 2011. “United States Life Tables, 2010.”

Social Security Administration. 2011. “Replacement Rates for Hypothetical Retired Workers.” SSA Publication No. 13-11747. Washington, DC: Social Security Administration.

Appendix

Literature

• UBI: insurance gains versus distortionary finance (Ghatak and Maniquet 2019; Conesa, Li, and Li 2023; Luduvice 2024)
• Alaska Permanent Fund evidence finds little labor-supply response; general-equilibrium effects may matter (Jones and Marinescu 2018)
• Future transfers can crowd out child investment and entrepreneurship (Daruich and Fernández 2024; Awal and Hashmi 2025)
• SWF research emphasizes fiscal stability rather than redistribution (Megginson and You 2014; James and MacDermott 2022)
• Corneo (2022) studies a redistributive SWF without the microeconomic financing channels here

[back]

Real-World “SWF”?

• Sovereign wealth funds generally invest globally for fiscal stabilization
• State-linked equity funds, such as Temasek, own domestic businesses
• Public investment banks invest in domestic firms
• Our SWF combines domestic business investment with citizen dividends

[back]

Household Problem

\[ V^o(a,h,\tilde z_o)=\max_{c,e,a'\ge0} \left\{u(c)+\beta\mathbb E[V(a',h',z')\mid h,e]\right\}, \] subject to \[ e\le y_o+(1+r)a-\tau(a,h,\tilde z)+\phi\omega, \] \[ c+e+a'\le y_o+(1+r)a-\tau(a,h,\tilde z)+\omega. \]

[back]

Entrepreneurial Production

\[ \max_\ell (\epsilon zk)^\alpha\ell^{1-\alpha}-w\ell, \qquad R(\epsilon z;w)=\epsilon z\alpha\left(\frac{1-\alpha}{w}\right)^{\frac{1-\alpha}{\alpha}}. \] \[ (1-\kappa)k(a,\epsilon z)\le\lambda(a+\phi\omega), \] \[ y(a,\epsilon z)=\max\{0,\lambda(a+\phi\omega)[R(\epsilon z;w)-r-\delta]\}. \]

[back]

Government Budget Constraint

\[ \begin{aligned} D&=\kappa\int R(\epsilon z;w)k(a,\epsilon z)\,dg(\epsilon)d\mu,\\ K_g&=\kappa\int k(a,\epsilon z)\,dg(\epsilon)d\mu, \end{aligned} \] \[ \omega+I_g+G_0=T_0+D,\qquad K_g'=(1-\delta)K_g+I_g. \] In steady state: \[ \omega+\delta K_g+G_0=T_0+D. \]

[back]

GE Channels

Direct policy effects and general-equilibrium feedbacks.
Solid arrows are partial-equilibrium channels.
Dashed arrows are equilibrium adjustments.

[back]

Fixed Parameters

Parameter	Description	Value	Source / target
\(\chi\)	CRRA coefficient	\(1.5\)
\(\varphi_1\)	Inverse Frisch elasticity	\(2.0\)	Fixed
\(q_1\)	Adult equivalence scale	\(2.057\)	Implied by calibrated altruism
\(\varsigma\)	Survival probability	\(0.577\)	Life expectancy, National Center for Health Statistics (2011)
\(\alpha\)	Entrepreneurial capital elasticity	\(0.33\)
\(\delta\)	Model-period depreciation rate	\(0.744\)
\(\lambda\)	Collateral constraint	\(2.00\)	Mean entrepreneur leverage, SCF
\(\psi_s\)	Earnings shifter	\((1,1.1403)\)	Hourly wage profile, CPS
\(\tau_1\)	Tax progressivity	\(0.20\)	Heathcote, Storesletten, and Violante (2017)
\(G/Y\)	Government purchases / GDP	\(18.9\%\)	Heathcote, Storesletten, and Violante (2017)
\(r^{ann}\)	Annual safe rate	\(3\%\)	Fixed empirical input
\(\omega_0/\bar y^{ann}\)	Benchmark universal transfer / worker earnings	\(4\%\)	Policy benchmark
\(\phi_{h0},\phi_{k0}\)	Benchmark transfer pledgeability	\((0.1327,0.1327)\)	Policy benchmark
\(p\)	Social Security replacement rate	\(40\%\)	Social Security Administration (2011)
\(\tau_a\)	Capital income tax rate	\(0.36\)	Fixed
\(\tau_c\)	Consumption tax rate	\(0.05\)	Fixed

[back]

Calibrated Parameters

Parameter	Description	Provisional	Target / role
\(\beta^{ann}\)	Discount factor	\(0.9895\)	Capital clearing at \(r^{ann}=3\%\)
\(\tau_0^{ann}\)	Tax level	\(0.9548\)	\(G/Y=18.9\%\)
\(\theta\)	Altruism toward children	\(0.4851\)	Transfers / wealth
\(\varphi_{0,s}\)	Labor disutility	\((7.0566,\;14.2298)\)	Mean hours by age
\(\zeta_s\)	Entrepreneurship preference	\((-1.1900,\;-0.5124,\;-1.0282)\)	Entrepreneur shares by age
\(\sigma_\xi\)	Taste-shock dispersion	\(0.20\)	Fixed provisionally; wealth-entry profile auxiliary
\(\gamma\)	Child investment productivity	\(0.4625\)	Education expenditures / GDP
\(\nu\)	Entrepreneurial returns to scale	\(0.9381\)	Entrepreneur wealth share
\(\mu_z\)	Entrepreneurial ability level	\(4.1987\)	Entrepreneurial income share
\(\rho_h\)	Intergenerational human-capital persistence	\(0.3875\)	Parent-child income rank slope
\(\rho_z\)	Human capital–entrepreneurial ability link	\(0.4500\)	Children’s relative entrepreneurship
\(\sigma_\eta^h,\sigma_\eta^z\)	Permanent skill dispersion	\((0.1583,\;0.1583)\)	Permanent earnings dispersion
\(\sigma_s^h\)	Labor-income shock dispersion	\((0.2511,\;1.0015)\)	Lifecycle earnings risk
\(\sigma_s^z\)	Entrepreneurial shock dispersion	\((0.6315,\;0.0999,\;0.6315)\)	Business-income dispersion
\(\bar y^{ann}\)	Earnings scale	\(2.4398\)	Mean worker earnings

[back]

Targeted Moments

Moment	Data	Model
Mean hours by age (CPS; model units)	\((0.281,\;0.227)\)	\((0.300,\;0.151)\)
Common worker-income component SD (DeBacker et al.)	\(1.05\)	\(5.355\)
Stage-specific worker-income innovation SD (DeBacker et al.)	\((0.877,\;1.04)\)	\((0.369,\;1.097)\)
Entrepreneur shares by age (SCF)	\((8.4\%,\;14.8\%,\;8.3\%)\)	\((0.18\%,\;14.67\%,\;0.85\%)\)
Entrepreneur wealth share (SCF)	\(45.2\%\)	\(8.31\%\)
Entrepreneur income share (SCF)	\(13.9\%\)	\(19.64\%\)
Common entrepreneur-income component SD (DeBacker et al.)	\(2.96\)	\(3.058\)
Stage-specific entrepreneur-income innovation SD, projection (DeBacker et al.)	\((3.08,\;4.09,\;4.88)\)	\((0.000,\;0.000,\;1.714)\)
Education expenditures / GDP (OECD)	\(5.8\%\)	\(7.04\%\)
Transfers / wealth (FEDS)	\(40\%\)	\(5.02\%\)
Parent-child income rank correlation Chetty et al. (2014)	\(0.341\)	\(0.603\)
Children’s relative entrepreneurship probability	\(1.60\)	\(0.934\)

Gray model entries were targeted in the full calibration design but were not re-targeted in the temporary expanded continuation.

[back]