The FDR quick-sale adjustment, explained with IRD's own example

FDR normally taxes 5% of what you held on 1 April and ignores your trades. But if you buy and sell the same FIF within one tax year, a quick-sale adjustment gets added on top: the lesser of 5% of the in-and-out shares at their average cost, and the actual gain you made on them. It can't go below zero, so a losing trade adds nothing. That's the whole rule; the rest is arithmetic, and IR461 includes a worked example that this guide computes line by line.

If you're still choosing between methods, FDR vs CV is the place to start. This guide assumes you know what FDR is and digs into its one genuinely fiddly corner.

Why the adjustment exists

Base FDR looks at a single date. Whatever your FIF holdings were worth on 1 April, you're taxed on 5% of it, and nothing you do during the year changes that number. Shares you buy in July have no opening value, so on the base calculation alone they'd produce no FIF income at all, even if you sold them in December for a fat profit.

That's the gap the quick-sale adjustment closes. Buy and sell the same FIF inside one tax year and the in-and-out portion gets taxed anyway, either at the usual 5% rate or at your actual gain, whichever is smaller.

When it triggers

The adjustment applies when your holding in the same FIF both increased and decreased in the same income year. Both movements are required:

• Only bought during the year? No adjustment. The new shares simply weren't in your 1 April value this year (they will be next year).
• Only sold during the year? No adjustment. You pay the full 5% on the opening value, even for shares you sold in the first week of April. More on that sting below.
• Bought and sold, same FIF, same year? The adjustment applies, whichever order the trades happened in.

It's checked per FIF. Buying Apple and selling Microsoft doesn't trigger anything; buying and selling Apple does.

The formula, piece by piece

The adjustment is the lesser of two amounts, floored at zero (s EX 52):

1. The peak holding amount. This is 5% × the peak holding differential × average cost, where:

• The peak holding differential is the lesser of (your greatest shareholding during the year minus your opening shareholding) and (your greatest shareholding minus your closing shareholding). In plain terms: how many shares passed through your hands without staying.
• The average cost is the total you paid across all purchases of that share during the year, divided by the total shares bought. Brokerage on the buys counts toward cost, which nudges the average up.

2. The actual gain. What you sold the quick-sold shares for, minus what they cost at that average price. Disposals count only to the extent they follow purchases made earlier in the year.

Taking the lesser of the two means IRD taxes your real profit when it's small and the standard 5% when the profit is large. The zero floor means a quick sale at a loss adds nothing (and takes nothing away from the base FDR either).

IR461's example, line by line

IR461 works through one full year of trading in "Company A". Here's the fact pattern:

Date	Action	Shares	Amount	Running holding
1 April	Opening balance	10,000	$200,000	10,000
1 October	Buy	5,000	$110,000	15,000
1 December	Sell	4,000	$100,000	11,000
23 December	Buy	2,000	$44,000	13,000
31 March	Closing value		$254,000	13,000

Now the five steps:

1. Peak holding differential. The greatest holding during the year was 15,000 shares. Opening was 10,000 and closing was 13,000, so the differential is the lesser of 5,000 and 2,000: 2,000 shares.
2. Average cost. Two buys: $110,000 + $44,000 = $154,000, for 5,000 + 2,000 = 7,000 shares. $154,000 ÷ 7,000 = $22 per share.
3. Peak holding amount. 5% × 2,000 × $22 = $2,200.
4. Actual gain. The December sale raised $100,000 for 4,000 shares that cost 4,000 × $22 = $88,000 at average cost. Gain: $12,000.
5. The adjustment. The lesser of $2,200 and $12,000 is $2,200.

Total FIF income for the year: 5% × $200,000 opening value = $10,000, plus $2,200, so $12,200.

Notice what the two limbs did there. The investor's real trading profit was $12,000, but the peak-holding formula capped the extra tax base at $2,200, because only 2,000 shares were genuinely in-and-out. Most of the December sale came out of the long-held opening position, and the base 5% already covers those.

This exact example is wired into the calculator's test suite, so what you see in the verdict for a quick-sale year is the same arithmetic IR461 prescribes, share counts, average costs and all.

Traps worth knowing

• The 2 April sale. Sell your whole holding two days into the tax year and base FDR still charges 5% of the full 1 April value. FDR doesn't prorate. In a year you're planning to exit, it's worth comparing CV before assuming.
• Mid-year buys are invisible this year, visible next. Shares bought in-year with no matching sale produce no FDR income now, then join the opening value on the next 1 April.
• Brokerage moves the numbers. Buy-side brokerage is part of the cost of the purchases, so it raises the average cost, which raises the peak holding amount slightly and lowers the actual gain. Small effect, but it's why your figures may not match a back-of-envelope version that ignores fees.
• Every trade needs an NZD value. Each buy and sell converts at the exchange rate for its own date, so the rate you use feeds the average cost and the gain.

Or just use CV for the year

Individuals and eligible trustees can choose between FDR and CV each year and take the lower result. CV measures your actual change in value, dividends included, with no separate quick-sale machinery, because your trades are already inside the calculation. In a heavy-trading year the two methods can land far apart, and there's no way to eyeball which one wins. Run both. That's what the calculator does on every portfolio, quick sales included, before it recommends a method.

In short

• The quick-sale adjustment only exists under FDR, and only when the same FIF was both bought and sold in the same tax year.
• It adds the lesser of the peak holding amount (5% × in-and-out shares × average cost) and your actual gain, never less than zero.
• IR461's example lands on $12,200: $10,000 of base FDR plus a $2,200 adjustment.
• Selling at a loss adds nothing, and selling early in the year doesn't reduce base FDR.
• CV has no quick-sale rule, which is one more reason to compare both methods each year.

Common questions

What triggers the quick-sale adjustment?

Buying and selling shares in the same FIF within the same tax year. Both movements are needed. A year with only purchases, or only sales, produces no adjustment, and trades in different FIFs don't interact.

I only sold shares this year. Does the adjustment apply?

No. Sales alone never trigger it. You still pay base FDR of 5% on the full 1 April market value of what you held, even if you sold everything early in the year, which is exactly the situation where checking CV is worthwhile.

What if my quick sale made a loss?

The adjustment is floored at zero. A loss on the in-and-out shares adds nothing to your FIF income, but it doesn't reduce the base 5% on your opening value either. FDR income for a holding can never be negative.

Is brokerage included in the average cost?

Yes. The average cost is built from what the in-year purchases actually cost you, brokerage included, per s EX 52(13). That slightly raises the average cost and lowers the calculated gain compared with using clean share prices.

Can I avoid the adjustment by using CV?

Individuals and eligible trustees can elect CV for a year where it gives a lower result, and CV has no quick-sale adjustment because it taxes your actual result directly. Whether CV actually comes out lower depends on the year, so calculate both before deciding.

Sources: IR461 (quick sale adjustment and the Company A worked example, p.15; FDR notes p.22), and section EX 52 of the Income Tax Act 2007.