Openai/693cbf4b-55bc-8002-9fea-d4feddfd1f55

This conversation was exported from ChatGPT

gpt-5-2

User: Why is this wrong

Why is this wrong

Hi, some Saturday thoughts.

If I sold my business for $1M and retired at age of 60, and needed immediate income until super kicks in.

I could buy $1M worth of QQQI with 13% yield ($130,000). US withholding tax is 15% ($19,500), so the after-tax dividend is $110,500.

Under FIF FDR, taxable income is $1M x 5% = $50,000. Effective tax rate is about 15%, so $7,500.

So with $1M, I could get about $103,000 after tax, is this correct?

Additionally, since I have already paid $19,500 WHT, can it be used to offset the $7,500 FIF income tax, since both taxes are on the same income?

Thanks!

Assistant: Short answer: it’s wrong in several important ways.

Short answer: it’s wrong in several important ways. The arithmetic looks tidy, but the tax logic doesn’t line up with how NZ actually treats this.

I’ll go point-by-point and call out the real pitfalls.

1. The 13% “yield” is not dividend income

QQQI’s distribution is not a clean 13% dividend. It’s an options-overlay fund. Distributions are typically a mix of:

• ordinary income
• capital gains
• return of capital (ROC)

Only the US-taxable portion gets US withholding. ROC is not income in the US sense and reduces cost base instead.

So this assumption is already shaky:

“13% yield = $130,000 of income”

It may be cash, but it is not necessarily taxable income, either in the US or NZ.

This matters a lot once FIF enters the picture.

2. FIF FDR ignores the actual yield entirely

Under FIF – FDR method, NZ does not care what QQQI actually pays you.

Taxable income is mechanically:

5% × opening market value

So yes, on $1M:

• FIF income = $50,000
• At ~15% marginal rate → $7,500 NZ tax

So far, that part is correct if you are using FDR and not CV.

But here’s the key consequence:

FIF tax is not a tax on dividends.

It is a deemed return on the asset value.

Which leads directly to the next problem.

3. US withholding tax does not offset FIF tax

This part is flatly incorrect:

“since I have already paid $19,500 WHT, can it be used to offset the $7,500 FIF income tax”

No. It cannot.

Why:

• Foreign tax credits are only allowed where the foreign tax is paid on the same income that NZ is taxing
• Under FDR, NZ is taxing a deemed 5% return, not the dividend
• The US withholding tax is paid on actual distributions
• IRD treats these as different income bases

So:

• The $19,500 WHT is effectively stranded
• It cannot be credited against FIF tax
• You still pay the full $7,500 in NZ

This is one of the most common (and expensive) misunderstandings with high-yield offshore ETFs.

4. Your “$103k after tax” result is internally inconsistent

Let’s line it up correctly.

Cash reality (approximate)

• Cash received: $130,000
• US WHT (only on taxable portion, not full amount): unknown, but assume ~$19,500 for argument
• Net cash in hand: ~$110,500

NZ tax reality

• FIF tax payable: $7,500
• No credit for WHT

True net cash

• $110,500 – $7,500 = ~$103,000

So your final number happens to land in the same place — but for the wrong reasons.

The danger is that:

• the WHT may be lower or higher than you think
• distributions can change
• ROC complicates future capital
• FIF tax stays fixed at 5% regardless

This is not a stable or tax-efficient structure.

5. The really big structural problem (the one that bites later)

If QQQI has a bad year and:

• price falls 20%
• distributions drop

You still owe:

5% of opening value in tax

Even if:

• cash yield falls to 6%
• or distributions are suspended

That’s the FDR asymmetry risk.

At age 60, drawing income, this is dangerous.

6. What people usually do instead (directionally)

Not prescribing, just pointing out reality:

• NZ PIE funds holding offshore assets
• Lower-yield / growth-tilted ETFs
• Blending NZ dividend income (no FIF) with offshore growth
• Managing taxable income deliberately to stay under thresholds

QQQI specifically is almost never optimal for NZ retirees once FIF is considered.

Bottom line

What’s wrong:

1. Treating “yield” as income
2. Assuming US WHT is creditable against FIF
3. Ignoring ROC mechanics
4. Ignoring FDR downside risk
5. Accidentally getting the right answer for the wrong reasons

One clarifying question (important before going further)

Do you expect to:

• spend the full distribution, or
• reinvest part of it?

That answer changes whether this structure is merely inefficient or actively dangerous.