A calibrated model of debt recycling with interest costs and tax shields: viability under different fiscal regimes and jurisdictions

TL;DR

This paper extends the debt-recycling framework by incorporating mortgage interest costs and tax shields, calibrating the model to Australia, Germany, and Switzerland for both owner-occupied and rental properties. By deriving the augmented average-dynamics matrix $\tilde{A}$ with eigenvalues $\tilde{\lambda}_{1,2}$, it maps phase boundaries across the parameter space defined by investment success probability $p$, housing drift $s$, and institutional parameters $(r_m, r_b, \tau_m, \tau_b)$. The results show that positive interest costs without tax shields reduce the viability of debt recycling, while tax shields partially offset those costs, with rental properties generally outperforming owner-occupied housing. Country-specific findings reveal systematic heterogeneity: Switzerland offers the most favorable conditions due to low rates and broad deductibility, Germany lies in between, and Australia presents the most challenging environment for owner-occupiers. The work has practical implications for households, lenders, and regulators, suggesting context-dependent policy tools such as dynamic LTV caps and differential tax treatment to manage leverage risks in debt-recycling arrangements, and highlighting the importance of cross-country differences in institutional settings.

Abstract

Debt recycling is a leveraged equity management strategy in which homeowners use accumulated home equity to finance investments, applying the resulting returns to accelerate mortgage repayment. We propose a novel framework to model equity and mortgage dynamics in presence of mortgage interest rates, borrowing costs on equity-backed credit lines, and tax shields arising from interest deductibility. The model is calibrated on three jurisdictions -- Australia, Germany, and Switzerland -- representing diverse interest rate environments and fiscal regimes. Results demonstrate that introducing positive interest rates without tax shields contracts success regions and lengthens repayment times, while tax shields partially reverse these effects by reducing effective borrowing costs and adding equity boosts from mortgage interest deductibility. Country-specific outcomes vary systematically, and rental properties consistently outperform owner-occupied housing due to mortgage interest deductibility provisions.

Figures (7)

• Figure 1: Calibrated Phase Diagrams for the Self-Owned Case in Australia, Germany and Switzerland with parameters $\ell = 0.5$, $\mu =0.5$, $E_0 = \$ 30,000$, $M_0 = \$ 300, 000$, $\pi^\star = \$3,000$, $q = 1\%$ and $(s, r_m, r_b, \tau_m, \tau_b)$ specified in Tab. \ref{['tab:calibration']}. Stars represent market conditions (housing-investment) in two representative years, namely Q4 2008 (financial crises, green) and Q2 2025 (current, red). The red dashed lines represent the range of the typical housing market fluctuations ($s$) observed in the jurisdiction.
• Figure 2: Quarterly calibration of ($p,s$) pairs for Australia, Germany, and Switzerland over the period 2010-2025. Each red star corresponds to the first quarter of a given year, with color intensity increasing over time to represent the temporal dimension of the calibration. The blue circle marks the starting point (Q1 2010), while the red circle denotes the most recent observation (Q1 2025). The green star corresponds to the global financial crisis (Q4 2008).
• Figure 3: Calibrated Simulation: Success Scenario with $p = 0.6$ and $s = 1.5\%$ for the Self-Owned Case.
• Figure 4: Calibrated Simulation: Default Scenario with $p = 0.4$ and $s = -1.5\%$ for the Self-Owned Case.
• Figure 5: Calibrated Phase Diagrams for the Rental Case in Australia, Germany and Switzerland with parameters $\ell = 0.5$, $\mu =0.5$, $E_0 = \$ 30,000$, $M_0 = \$ 300, 000$, $\pi^\star = \$3,000$, $q = 1\%$ and $(s, r_m, r_b, \tau_m, \tau_b)$ specified in Tab. \ref{['tab:calibration']}. Stars represent market conditions (housing-investment) in two representative years, namely Q4 2008 (financial crises, green) and Q2 2025 (current, red). The red dashed lines represent the range of the typical housing market fluctuations ($s$) observed in the jurisdiction.