Extensions to the Wealth Tax Neutrality Framework

This paper extends Froeseth's (2026) wealth tax neutrality framework by demonstrating that while neutrality holds under stochastic volatility and Epstein-Zin preferences, it breaks under non-homothetic preferences and four specific implementation channels—such as progressive thresholds and inelastic markets—which are formalized and calibrated to the Norwegian system to evaluate global minimum tax proposals.

The Gist

Imagine you are the captain of a massive ship (your investment portfolio) sailing through the ocean of the financial markets. Your goal is to reach a destination (your retirement or future wealth) as efficiently as possible.

This paper asks a simple but profound question: If the government puts a toll booth on your ship that charges a flat fee based on the size of your boat, does it change how you steer the ship?

The author, Anders Frøseth, builds on a previous idea that said: "No, it doesn't." If the tax is fair and applies to everything equally, you should keep sailing the same course. But this paper digs deeper to see if that holds true when the ocean gets stormy, when your captain's personality changes, or when the toll booth is designed poorly.

Here is the breakdown of the paper using simple analogies:

Part 1: When the "No Change" Rule Still Works

The paper first checks if the "neutral" result holds up in more realistic, messy scenarios.

• The Stormy Ocean (Stochastic Volatility): In the old model, the ocean was calm and predictable. In reality, waves get huge and unpredictable (volatility). The paper finds that even if the waves are chaotic, as long as you are a rational captain who cares about your wealth proportionally (CRRA preferences), the toll booth doesn't make you change your steering. You still sail the same route; you just arrive with slightly less fuel (wealth) because of the toll.
• The "Time-Traveling" Captain (Epstein-Zin Utility): Some captains care differently about risk vs. time. The paper shows that even with these complex preferences, the toll booth still doesn't change your steering.
• The "Survival Mode" Captain (HARA Preferences): Here is where the rule breaks. Imagine a captain who is terrified of running out of food (subsistence needs). If a tax takes away some of their fuel, they panic. They stop taking risks and steer toward the safest, slowest route to ensure they don't sink. The tax changes the course because the captain is scared.

Part 2: Why Real-World Taxes Do Change the Course

The paper argues that in the real world, wealth taxes are rarely "fair" or "uniform." They have four main flaws that force captains to change their steering:

1. The "Discounted Ticket" Problem (Non-Uniform Assessment)
Imagine the toll booth charges full price for your cargo of gold (stocks) but gives you a 75% discount for your cargo of houses.

• The Result: You stop buying gold and start buying houses, not because houses are better, but because the tax man likes houses more.
• Real World: In Norway, your primary home is taxed at a tiny fraction of its value, while your bank savings are taxed at full value. This tricks people into holding too many houses and not enough stocks.

2. The "Squeezed Market" Problem (Inelastic Markets)
Imagine a crowded dance floor. If a few people leave, the music keeps playing. But what if the dance floor is packed so tight that if everyone tries to leave at once, the doors jam and the floor collapses?

• The Result: When a tax forces wealthy people to sell assets to pay the bill, they might crash the market prices. It's not just that they have less money; it's that the value of everyone's assets drops because the market can't absorb the sudden selling.
• Real World: If billionaires are taxed heavily, they might sell stocks. Because there aren't enough buyers ready to step in, stock prices could drop sharply, hurting even people who didn't pay the tax.

3. The "Cliff Edge" Problem (Progressive Thresholds)
Imagine a toll booth where the first $1 million is free, but the moment you cross $1.000.01, you pay a huge fee on the entire amount.

• The Result: This creates a weird incentive. If you are just above the line, you might take more risks. Why? Because the "free" part of your wealth acts like a safety net (a tax shield). It's like having a parachute that lets you jump out of a plane with more confidence.
• The Twist: This is the opposite of the "Survival Mode" captain. Here, the tax makes people bolder near the threshold, but scared if they are poor.

4. The "Run Away" Problem (Migration)
If the toll booth is too expensive, the captain might just sail to a different country where there is no toll.

• The Result: If the tax is too high, the wealthy don't just change their portfolio; they change their address. This is a "participation margin."
• Real World: The paper looks at Norway after 2022. When taxes went up, many wealthy Norwegians moved to Switzerland. It's like a game of musical chairs where the wealthy players just leave the table if the rules get too tough.

Part 3: The Big Proposals (Saez-Zucman vs. France)

The paper tests two famous ideas for taxing the ultra-rich:

• The Global Plan (Saez-Zucman): A 2% tax on billionaires.
  • Verdict: This is actually quite "clean." Because it targets only the super-rich (who don't need a safety net) and taxes everything at market value (no discounts), it avoids most of the steering problems. The main risk is just the "Squeezed Market" effect if they all sell at once.
• The French Plan: A 2% tax on the rich (starting at €100 million).
  • Verdict: This is messier. Because the threshold is lower, it hits people who are closer to the "Cliff Edge." It creates more incentives for them to take weird risks or move to other countries.

The Bottom Line

The paper concludes that tax neutrality is a fragile thing.

If you design a wealth tax perfectly (flat rate, no discounts, global coordination), it might not change how people invest. But in the real world, taxes are messy. They have discounts for houses, they have "cliff edges," and they have loopholes.

The Analogy Summary:
Think of the ideal tax as a flat fee for driving on a highway. It doesn't matter if you drive a Ferrari or a truck; you pay the same per mile. You drive the same speed.
But a real-world wealth tax is like a toll system where:

• Trucks get a discount.
• If you drive too fast, you pay double.
• If you cross a certain bridge, you pay a massive fee.
• If you don't like the toll, you can just drive to a different country.

The paper shows that these "real world" rules force drivers to change their speed, their route, and sometimes, their entire destination.

Technical Summary

Technical Summary: Extensions to the Wealth Tax Neutrality Framework

1. Problem Statement

The paper investigates the robustness of the wealth tax neutrality result established in Froeseth (2026). The original framework demonstrated that a proportional wealth tax on market values is neutral regarding portfolio choice, Sharpe ratios, and equilibrium prices under Constant Relative Risk Aversion (CRRA) preferences and Geometric Brownian Motion (GBM).

This paper addresses two critical gaps:

1. Theoretical Robustness: Does neutrality hold under more complex, empirically relevant market dynamics (stochastic volatility) and more general utility specifications (Epstein-Zin, HARA)?
2. Real-World Deviations: Through which specific institutional channels do implemented wealth taxes (specifically the Norwegian system and proposed global variants) depart from neutrality, even when investors possess CRRA preferences?

2. Methodology

The paper employs a continuous-time dynamic optimization framework (Merton problem) extended to various stochastic environments and preference structures.

• Modeling Framework:
  • Stochastic Volatility: Utilizes the Heston model and general Markov diffusions to model time-varying variance and leverage effects.
  • Utility Functions: Analyzes CRRA, Hyperbolic Absolute Risk Aversion (HARA), and Epstein-Zin recursive utility.
  • Tax Structures: Models proportional taxes, asset-class-specific assessment discounts (valuation fractions), progressive thresholds (brackets), and endogenous labor supply.
  • General Equilibrium: Incorporates the "Inelastic Markets Hypothesis" (Gabaix & Koijen, 2021) to analyze price impacts of tax-induced flows.
• Analytical Approach:
  • Derives Hamilton-Jacobi-Bellman (HJB) equations to solve for optimal portfolio weights ($w^*$) and consumption.
  • Utilizes homogeneity properties of value functions to determine tax invariance.
  • Calibrates models using Norwegian tax parameters (2026 rates, assessment fractions) and empirical data from the US, Switzerland, and Colombia.
  • Applies comparative statics to evaluate the Saez-Zucman global proposal and the French national minimum tax.

3. Key Contributions and Results

A. Extension of the Neutrality Frontier

The paper establishes the precise conditions under which neutrality holds or breaks:

1. Stochastic Volatility (Propositions 1 & 2):

   • Result: Neutrality is preserved under the Heston model and general Markov diffusions.
   • Mechanism: Under CRRA, the value function is homogeneous of degree $1-\gamma$ in wealth. This homogeneity ensures that optimal portfolio weights (including Merton's intertemporal hedging demand) are independent of the wealth level. Since a proportional tax scales wealth but does not alter the relative risk-return trade-offs or the volatility state dynamics, the optimal portfolio composition remains unchanged.
   • Significance: Neutrality extends beyond GBM to models capturing fat tails and leverage effects, provided preferences are CRRA.

2. Preference Specifications (Propositions 3 & 4):

   • Epstein-Zin Utility: Neutrality is preserved. Separating risk aversion ($\gamma$) from the elasticity of intertemporal substitution ($\psi$) does not break the homogeneity of the value function.
   • HARA Preferences: Neutrality breaks. HARA utility includes a subsistence consumption parameter ($\zeta$), creating a "floor wealth" ($H$). A wealth tax reduces total wealth, increasing the ratio of floor wealth to surplus wealth ($H/W$), which effectively raises the investor's relative risk aversion.
   • Result: Under HARA, the optimal risky share decreases as the tax rate increases ($\Delta w^* < 0$). This distortion is regressive, affecting lower-wealth investors most severely.

B. Channels of Non-Neutrality (Real-World Deviations)

Even with CRRA preferences, four institutional channels break neutrality:

1. Non-Uniform Assessment (Proposition 5):

   • Mechanism: When tax authorities apply valuation discounts (e.g., Norway's 20% discount on shares vs. 100% on deposits), the effective tax rate differs across asset classes.
   • Result: Investors tilt portfolios toward assets with lower assessment fractions ($\alpha_i < \alpha_0$).
   • Calibration: In Norway, the 75% discount on primary housing creates a massive portfolio tilt (approx. +18.75 percentage points) toward housing relative to financial assets, distorting capital allocation.

2. Inelastic Markets (General Equilibrium Effects):

   • Mechanism: Wealth taxes reduce investor wealth, forcing aggregate selling. Under the Inelastic Markets Hypothesis (multiplier $M \approx 5$), these flows cause price declines exceeding fundamental valuation changes.
   • Result: The tax is not neutral regarding equilibrium prices. Prices fall by $\Delta P/P \approx -M \cdot \tau_w \cdot \phi$, where $\phi$ is the share of market cap held by taxed investors. This shifts the tax burden to non-taxed investors.

3. Progressive Thresholds (Proposition 6):

   • Mechanism: A progressive tax with an exemption threshold creates a "tax shield" (the present value of the exemption).
   • Result: Contrary to HARA, the tax shield acts as a riskless endowment, increasing the optimal risky share ($\Delta w^* > 0$) for investors near the threshold.
   • Interaction: This effect opposes the HARA distortion. For moderate wealth, the tax shield may partially offset the subsistence floor effect, but for the ultra-wealthy, the distortion vanishes.

4. Endogenous Labour Supply (Proposition 7):

   • Result: Under proportional taxation, portfolio choice and labor supply are separable; the tax only induces an income effect (increasing labor supply). However, at progressive thresholds, a "notch" creates a substitution effect that may reduce labor effort or entrepreneurial activity to stay below the tax bracket.

C. Tax-Induced Migration (Section 9)

• Mechanism: Under domicile-based taxation, investors migrate if the present value of future tax liabilities exceeds migration costs.
• Result: A progressive threshold creates a participation margin where investors exit the jurisdiction entirely.
• Case Study (Norway 2022-2024): Following tax rate hikes and the abolition of the exit tax expiration, high-wealth emigration spiked. However, the paper notes the difficulty in isolating the wealth tax effect from concurrent changes in dividend taxes and exit tax rules.

4. Policy Applications and Evaluations

The framework is applied to evaluate two major proposals:

1. Saez-Zucman Global Minimum Tax (2% on >$1B):

   • Neutrality Profile: High neutrality. The high threshold eliminates progressive distortions; market-value assessment eliminates non-uniformity; CRRA is a valid approximation for billionaires.
   • Primary Risk: Inelastic market effects. While the flow impact is modest globally (~0.8% price decline), it is non-negligible.
   • Migration: Requires global coordination to prevent jurisdiction shopping.

2. French National Minimum Tax (2% on >€100M):

   • Neutrality Profile: Lower neutrality. The lower threshold activates the progressive tax shield (increasing risk-taking) for a broader population.
   • Risk: High migration risk due to the "national" nature of the tax within the EU single market, potentially triggering capital flight.

3. Norwegian System:

   • Neutrality Profile: Low neutrality. Severely distorted by valuation discounts (housing vs. shares), progressive thresholds, and debt deductibility rules.

5. Significance and Implications

• Theoretical: The paper delineates the boundary of wealth tax neutrality. It proves that neutrality is robust to complex market dynamics (stochastic volatility) but fragile to non-homothetic preferences (HARA) and institutional design flaws (non-uniform assessment).
• Policy Design:
  • Uniformity is Key: To preserve neutrality, wealth taxes must apply uniformly across asset classes at market value. Valuation discounts, while politically popular or administratively convenient, create significant capital misallocation.
  • Thresholds: Progressive thresholds create a "tax shield" that encourages risk-taking near the exemption boundary, a counter-intuitive effect that must be weighed against distributional goals.
  • Coordination: Unilateral national wealth taxes face severe migration constraints. Global coordination (as in the Saez-Zucman proposal) is necessary to eliminate the migration margin, though it introduces general equilibrium price risks in inelastic markets.
• Empirical: The paper provides a roadmap for testing these channels using registry data (e.g., Norwegian wealth registry) and highlights the importance of distinguishing between short-run portfolio inertia and long-run equilibrium distortions.

In conclusion, while a theoretically pure proportional wealth tax on market values is neutral under CRRA, real-world implementation almost invariably introduces distortions through assessment rules, market inelasticity, and progressive structures. The magnitude and direction of these distortions depend critically on the specific design of the tax and the wealth level of the investor.