📈 economics

Flow Taxes, Stock Taxes, and Portfolio Choice: A Generalised Neutrality Result

This paper demonstrates that a comprehensive system of ownership taxes preserves portfolio neutrality by acting as a uniform drift shift and rescaling of wealth dynamics, provided that corporate and capital income tax rates are aligned, shielding mechanisms match the risk-free rate, and wealth tax assessments are uniform, with deviations from these conditions generating separable distortions where non-uniform wealth taxation dominates flow-tax effects.

Original authors: Anders G Frøseth

Published 2026-03-18

📖 8 min read🧠 Deep dive

CC BY 4.0

Original authors: Anders G Frøseth

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: The "Tax Maze"

Imagine you are an investor trying to build the perfect garden (your investment portfolio). You have different types of plants: risky ones like exotic flowers (stocks) and safe ones like sturdy oak trees (bonds or cash).

Your goal is to pick the best mix of plants to get the most beauty (return) for the least amount of worry (risk).

Now, imagine the government puts up a fence around your garden and charges you fees. The paper asks a simple question: Do these fees change which plants you choose to grow?

The author, Anders Frøseth, argues that if the tax system is designed correctly, the answer is no. You will still choose the exact same mix of plants, even if the government takes a chunk of your harvest. However, if the tax rules are messy or unfair, you will start picking the wrong plants just to avoid the fees.

The Two Types of Taxes: The "Flow" vs. The "Stock"

To understand the paper, we need to distinguish between two ways the government taxes your garden:

Flow Taxes (The "Harvest Tax"): These are taxes on the growth or income your plants produce.
- Examples: Corporate tax (tax on company profits), taxes on dividends (cash payouts), and taxes on capital gains (profit when you sell a plant).
- Analogy: Imagine the government takes 30% of every apple your tree produces. It doesn't matter if the tree is big or small; they just take a slice of the fruit.
Stock Taxes (The "Garden Size Tax"): This is a tax on the total value of your garden, regardless of whether it produced fruit this year.
- Example: Wealth tax.
- Analogy: Imagine the government charges you $100 just for owning a garden that is 1,000 square feet. It doesn't matter if the plants are dead or alive; you pay based on the size of the land.

The Magic Trick: How Neutrality Works

The paper uses a complex math tool called the Fokker–Planck equation (think of it as a weather map for your money). It shows how your wealth moves up and down over time.

The author discovers a "Magic Symmetry" that happens when three specific conditions are met. When these conditions hold, the taxes act like a uniform elevator or a global dimmer switch:

The Corporate Tax matches the Personal Tax: The tax on company profits is the same as the tax on your personal savings.
The "Shielding" Rate matches the Safe Rate: The government allows you to earn a "safe" return (like a bank deposit) tax-free before taxing the extra profit. This safe rate must match the actual risk-free interest rate.
The Wealth Tax is Fair: The tax on your garden size is calculated exactly the same way for every type of plant.

What happens when these rules are followed?

The Dimmer Switch (Flow Taxes): The flow taxes turn down the brightness of all your potential profits by the exact same percentage. If your risky flowers usually grow 10% and your safe trees grow 2%, the taxes might cut both to 6% and 1%. The difference between them stays the same. You still prefer the flowers over the trees for the same reasons.
The Elevator (Stock Taxes): The wealth tax lowers the floor of your garden by a fixed amount for everyone. It shifts your total wealth down, but it doesn't change the slope of the hill. You are still climbing the same hill, just starting from a lower point.

The Result: You don't change your portfolio. You don't sell your risky stocks to buy bonds just to avoid taxes. The "Tangency Portfolio" (the perfect mix of risk and reward) remains untouched.

When the Magic Breaks: The "Distortions"

The paper explains what happens when the tax rules are messy. This is where the "elevator" breaks and starts tilting the floor.

1. The "Uneven Floor" (Violation of Condition 3)

This is the biggest problem in the real world (specifically in Norway).

The Scenario: The government taxes your garden size, but they value different plants differently.
- Bank deposits: Valued at 100% (You pay full tax).
- Listed Stocks: Valued at 80% (You pay less tax).
- Holiday Homes: Valued at 30% (You pay very little tax).
The Analogy: Imagine the government says, "We will charge you rent based on the size of your garden, but we only count 30% of your holiday home's size, while counting 100% of your bank account."
The Result: You will stop planting flowers and start building holiday homes, not because they are better investments, but because they are tax cheaper. This is a "portfolio tilt." The paper calculates that in Norway, this distortion is 300 times stronger than any other tax error.

2. The "Broken Shield" (Violation of Condition 2)

The Scenario: The "safe" return you are allowed to earn tax-free is set too high or too low compared to the actual risk-free rate.
The Analogy: Imagine the government says, "You can keep the first $5,000 of your garden's growth tax-free." But the actual safe growth rate of the market is only $3,000.
The Result: You might shift your money slightly between stocks and bonds, but you won't change the mix of your risky stocks. It's a small wobble, not a collapse.

3. The "Double Standard" (Violation of Condition 1)

The Scenario: The tax on company profits is different from the tax on your personal savings.
The Result: Similar to the broken shield, this creates a tilt between stocks and bonds, but it doesn't mess up your choice of which stocks to buy.

The "Norwegian Miracle" and the "Book Value" Trap

The paper looks at Norway's specific tax system (the Aksjonærmodellen).

The Good News: Norway is actually very good at the "Magic Symmetry." The corporate tax and personal tax rates are identical (22%), and the "shielding" rate is set very close to the real risk-free rate. This means the flow taxes are perfectly neutral.
The Bad News: The wealth tax assessment is messy.
- Listed Stocks: Taxed on 80% of their market value.
- Unlisted Stocks (Private Companies): Taxed on 80% of their book value (accounting value).
- The Trap: For many modern tech companies, the "book value" is tiny compared to their "market value" (because their value is in ideas, not factories). So, the government taxes them on a tiny fraction of their real worth.
- The Consequence: Investors are heavily incentivized to put money into private, unlisted companies and owner-occupied houses, and avoid bank deposits and listed stocks. This distorts the economy, pushing capital into tax-favored assets rather than the most efficient ones.

The "Payment" Problem

The paper also notes that paying the wealth tax is expensive because of the flow taxes.

If you need cash to pay your wealth tax, you might sell stocks. But selling stocks triggers a capital gains tax.
Or, you might ask your company for a dividend. But the company pays corporate tax, and then you pay dividend tax.
The Analogy: It's like trying to fill a bucket with a hole in it. To get $1 into your pocket to pay the tax, you might need to generate $2 or $3 of pre-tax profit because of the "leaks" (taxes) along the way. This makes paying the tax painful and forces people to sell assets they didn't want to sell.

The Takeaway

Tax Neutrality is Possible: If you design a tax system where flow taxes (income) and stock taxes (wealth) are uniform and fair, investors will keep making the best economic choices, regardless of the tax rate.
The "Shielding" Deduction is a Hero: The rule that lets you earn a "normal" return tax-free is crucial. It stops the tax system from treating safe and risky investments unfairly.
The Real Enemy is Uneven Valuation: The biggest distortion doesn't come from how much tax you pay, but how you value the assets. If the government values your house differently than your stocks, you will move your money to the house, even if the house is a worse investment.
Norway's Lesson: Norway has a great flow-tax system, but its wealth tax assessment is the "Achilles' heel." Fixing the valuation discounts (making them uniform) would do more to fix the economy than tweaking the tax rates.

In short: The paper proves that if you tax everyone's garden fairly and uniformly, the garden will grow naturally. But if you tax the roses differently than the tulips, people will stop planting roses, and the garden will look very different than it should.

1. Problem Statement

The paper addresses a fundamental question in public finance and asset pricing: Does a complex, multi-layered tax system distort an investor's optimal portfolio composition?

While the neutrality of a proportional wealth tax (a tax on the stock of wealth) is well-established in literature (acting as a uniform drift shift), real-world investors face a combination of "flow taxes" (taxes on returns) and "stock taxes." Specifically, the paper analyzes the Norwegian dual income tax system (the aksjonærmodellen), where shareholders face four distinct layers of taxation:

Corporate Tax ( $\tau_c$ ): Levied on gross profits.
Capital Income Tax ( $\tau_k$ ): Levied on the risk-free component of returns (shielded portion).
Dividend/Capital Gains Tax ( $\tau_d$ ): Levied on the excess return above the shielding rate.
Wealth Tax ( $\tau_w$ ): Levied on net assets, often with asset-class-specific valuation discounts.

The core problem is determining whether the interaction of these flow and stock taxes preserves portfolio neutrality (invariance of optimal portfolio weights) or introduces distortions that tilt asset allocation.

2. Methodology

The paper employs a stochastic control framework based on the Fokker–Planck equation for wealth dynamics, extending the work of Frøseth (2026d) and Bouchaud & Mézard (2000).

Wealth Dynamics: The investor's log-wealth ( $x = \ln W$ ) follows a diffusion process $dx = v dt + \sigma_P dB$ , where $v$ is the drift velocity and $\sigma_P$ is the volatility.
Drift-Shift Symmetry: The author defines neutrality as an invariance property. A tax is neutral if it modifies the drift velocity $v$ $v$ in a way that does not alter the relative drifts between assets.
- Stock Tax: Acts as a uniform shift ( $v \to v - \tau_w$ ).
- Flow Taxes: Act as modifications to the return channel.
Generalized Transformation: The paper introduces a "Drift-Shift-and-Rescale" transformation. It analyzes how the combined tax system modifies the drift vector $v$ and the diffusion coefficient $D$ .
Assumptions:
- Geometric Brownian Motion (GBM) for asset returns.
- Constant Relative Risk Aversion (CRRA) preferences.
- Continuous time.
- The analysis focuses on the tangency portfolio (the composition of risky assets) rather than the total allocation to risky vs. risk-free assets.

3. Key Contributions

The paper makes three primary theoretical contributions:

Generalized Neutrality Theorem: It proves that a complex system of flow and stock taxes preserves portfolio neutrality if and only if three specific conditions are met. This generalizes the "drift-shift" result of pure wealth taxes to a "drift-shift-and-rescale" result.
Additive Separability of Distortions: It demonstrates that distortions arising from flow taxes (violations of neutrality conditions) and stock taxes (non-uniform assessment) are additively separable. They do not interact; the presence of a wealth tax does not amplify or dampen flow-tax distortions, and vice versa.
The Shielding Deduction as a Symmetry Restorer: It identifies the "shielding deduction" (exempting the risk-free return from the higher dividend tax rate) not just as an equity mechanism, but as a critical efficiency mechanism. It restores the symmetry between equity and debt taxation, ensuring the tax system acts as a uniform rescaling rather than a distortionary wedge.

4. Key Results

A. The Three Conditions for Neutrality

The combined tax system preserves optimal portfolio weights ( $w^*$ ) if and only if:

(C1) Rate Equality: The capital income tax rate equals the corporate tax rate ( $\tau_k = \tau_c$ ).
(C2) Shielding Alignment: The shielding rate (the threshold for the lower tax rate) equals the risk-free rate ( $r_s = r_f$ ).
(C3) Uniform Assessment: The wealth tax assessment fraction is uniform across all asset classes ( $\alpha_i = \alpha$ ).

B. The Mechanism of Neutrality

When (C1)–(C3) hold, the after-tax excess return vector is:
$R^{ex}_{after} = (1-\tau_c)(1-\tau_d)(\mu - r_f \mathbf{1}) + \text{constant}$

Flow Taxes: Uniformly rescale the excess drifts by the factor $k = (1-\tau_c)(1-\tau_d)$ . This contracts the efficient frontier but preserves the tangency portfolio (the direction of optimal risky weights).
Wealth Tax: Uniformly shifts all drifts by $-\tau_w \alpha$ . This shifts the opportunity set vertically but does not change relative drifts.
Result: The optimal portfolio weights remain identical to the pre-tax case.

C. Classification of Distortions (Symmetry Breaking)

When conditions fail, specific distortions arise:

Violation of (C1) or (C2): Distorts the equity-debt split (tilt along the capital allocation line) but preserves the tangency portfolio among risky assets.
Violation of (C3): Distorts the tangency portfolio itself. Non-uniform assessment fractions ( $\alpha_i \neq \alpha_j$ ) create an asset-specific drift shift (anisotropic field), causing investors to tilt portfolios toward assets with lower assessment fractions (e.g., housing or unlisted shares).

D. Quantitative Calibration (Norwegian System)

Calibrated to the Norwegian tax system:

Conditions (C1) and (C2) hold by design: $\tau_k = \tau_c = 22\%$ , and the shielding rate tracks the risk-free rate. Thus, flow taxes are neutral.
Condition (C3) is violated: Asset classes have different valuation discounts (e.g., bank deposits $\alpha=1.0$ , primary housing $\alpha=0.25$ , listed shares $\alpha=0.80$ ).
Magnitude: The distortion from non-uniform assessment is roughly 300 times larger than any residual distortion from flow-tax misalignment.
Impact: The system creates a massive incentive to hold owner-occupied housing and unlisted shares (assessed at book value) over bank deposits and listed shares.

5. Significance and Implications

Policy Design: The paper provides a formal efficiency argument for the shielding deduction. While often justified on equity grounds (protecting "normal" returns), this paper shows it is essential for allocative neutrality. Without it, the tax system would distort the equity-debt choice.
Tax Reform Priorities: The analysis suggests that reforming flow tax rates (corporate or dividend rates) has negligible impact on portfolio composition compared to reforming wealth tax assessment rules. Equalizing assessment fractions across asset classes would yield far greater allocative efficiency than adjusting tax rates.
Theoretical Unification: The paper unifies the literature on wealth taxes and flow taxes. It shows that the "drift-shift" symmetry of wealth taxes and the "rescaling" symmetry of flow taxes can coexist, provided the tax structure is symmetric.
Interaction with Market Frictions: The paper notes that while the composition is neutral, flow taxes amplify the cost of paying the wealth tax (e.g., via dividend extraction or share sales), effectively doubling the gross earnings required to pay a unit of wealth tax. This amplifies liquidity frictions and market impact costs, even if the optimal portfolio weights theoretically remain unchanged.

In conclusion, Frøseth demonstrates that a multi-layered tax system can be neutral, but this neutrality is fragile and relies on specific institutional designs (rate equality and shielding alignment). In the Norwegian case, the system is largely neutral regarding flow taxes, but the non-uniform wealth tax assessment remains the single dominant source of portfolio distortion.