A £400k buy-to-let nets £4,800 a year — the same cash in Segro pays £17,000 inside an ISA

Start with the gross. £400,000 flat, 6% gross yield is £24,000 a year. That figure flatters everything that follows, because gross yield is a gross misrepresentation of what landlords actually keep.

Deduct the obvious: 5% void allowance (£1,200), 10% letting agent management (£2,400), service charge and ground rent on a typical London or Manchester flat (£2,000), buildings insurance and gas safety certificates (£500), repairs and maintenance reserve at 1% of property value (£4,000). That leaves £13,900 of pre-mortgage, pre-tax operating profit. The mortgage on a £300,000 BTL loan at 5.5% is £16,500 of interest a year — an interest-only BTL, which is what most landlords run.

At this point a pre-2017 landlord would deduct mortgage interest as an expense and pay tax on the £-2,600 loss, getting a relief refund. After Section 24's 2020 finalisation, mortgage interest is no longer an expense — it becomes a 20% tax credit. So a higher-rate landlord adds the £16,500 of interest back, pays 40% tax on £13,900 of "profit" (£5,560), and then claws back 20% of the £16,500 interest as a credit (£3,300). Net tax bill: £2,260. Net income after tax: £-2,600 + £3,300 - £2,260 + £6,160 ≈ £4,800.

A basic-rate landlord does better — closer to £8,000 net — because Section 24 hurts higher-rate payers most. But the median UK BTL landlord is a higher-rate taxpayer (HMRC's own landlord population data shows about 60% of rental income flows to people earning above the higher-rate threshold). The £4,800 figure is the case the BTL industry doesn't put in the brochure.

Buy-to-let starts with a £20,000 hole on a £400,000 purchase. Standard SDLT on a £400,000 purchase as a primary residence is £10,000 (£2,500 at 2% on the £125k-£250k slice, plus £7,500 at 5% on the £250k-£400k slice, per the gov.uk SDLT table). Add the 5% additional-property surcharge on the full purchase price — £20,000 — and the buy-to-let buyer hands £30,000 to HMRC in stamp duty before turning the key.

That's a 7.5% transaction cost on the property — nearly two years of Segro dividend at 4.46%, paid before you've collected a single month's rent. And it's not a one-off you can amortise away cleanly: it sits in your basis cost for CGT purposes, which means it's only "recovered" if the property appreciates enough to offset it on disposal — which then triggers more tax.

The exit is where the second hammer falls. UK residential CGT for higher-rate payers is now 24% (from 28% pre-October 2024 — this is the lower of the two recent residential CGT rates). The annual exempt amount is £3,000. So if your £400,000 BTL has compounded at 3% a year for 10 years, it's now worth £537,500 — a £137,500 gross gain. After deducting the £30,000 SDLT, agent and legal fees on the way out (£10,000), and the £3,000 annual exemption, the taxable gain is around £94,500. CGT at 24% takes £22,700.

Run the same growth assumption on Segro inside an ISA: zero CGT on disposal, ever. The ISA wrapper is the most under-rated tax shelter in UK personal finance and BTL deliberately cannot use it.

A 2026 stocks-and-shares ISA gives a higher-rate taxpayer a £20,000 annual allowance (per the HMRC ISA allowance schedule) — fill it for five years and you have £100,000 inside the wrapper. The same £100,000 can be split across four UK REITs to get instant diversification across thousands of physical properties:

• Segro (SGRO.L), 697p, 4.46% dividend yield, P/E 17.0, market cap £9.4bn — industrial logistics warehouses, the rails of UK e-commerce.
• British Land (BLND.L), 385p, 5.94% dividend yield, P/E 8.6, market cap £3.95bn — London offices and retail parks.
• LondonMetric Property (LMP.L), 190p, 6.41% dividend yield, P/E 12.7, market cap £4.4bn — long-let logistics and convenience retail.
• Primary Health Properties (PHP.L), 94p, 7.74% dividend yield, P/E 13.5, market cap £2.45bn — NHS GP surgeries on government-backed leases.

Blend them equally and the portfolio yield is roughly 6.1% — every penny of which is income-tax-free inside the ISA, and every penny of capital growth on top is CGT-free forever. That's £6,100 of guaranteed dividend income on £100,000, plus capital appreciation, plus the right to sell any morning the market is open.

The stock yields above are point-in-time and will move. Land Securities' headline 12.89% trailing yield is distorted by special distributions and is not a reliable forward number — that's exactly the sort of figure a REIT investor must check at source. But across the basket, a 5-7% income yield with single-stock risk diversified across roughly 9,000 underlying tenanted properties is a structurally better income proposition than a single £400,000 flat in zone 4.

On liquidity: a Segro position can be sold by 9.05am on a working day. A buy-to-let flat takes three to nine months to sell, with estate agent fees of 1.5%, legal fees of £1,500, and an EPC requirement that may force a £5,000-£15,000 retrofit before you're allowed to list. The illiquidity discount on direct property is real — and it's a discount, not a premium.

The pro-BTL case rests almost entirely on leverage. Borrow 75%, multiply your equity exposure 4x, and small capital growth percentages compound into life-changing gains. The argument is real, but it cuts both ways and the cost has gone up.

A 75% LTV buy-to-let mortgage at 5.5% costs you £16,500 a year on a £300,000 loan. A 3% per-year property growth rate gives you £12,000 of capital appreciation in year one. You're paying £16,500 of interest for £12,000 of growth before tax. The leverage works only when capital growth is materially above the mortgage rate — and UK average house prices have grown around 4.0% per year over the last 10 years per ONS HPI, barely covering today's BTL borrowing cost.

When the Bank of England base rate sat at 0.1% in 2021, BTL leverage was a money-printing machine. With base rate now at 3.75% (30 April 2026), and BTL mortgage rates anchored at 5%+, the spread has collapsed. Leverage cuts your downside protection too: a 10% house-price fall on a 75% LTV portfolio wipes out 40% of your equity. The 2007-08 BTL collapse wasn't ancient history — it was the same maths.

REITs are also leveraged. Segro's loan-to-value ratio sits around 30%, British Land's around 35% — but at corporate borrowing rates well below retail BTL rates, and managed by treasury teams whose day job is hedging. You get a more conservatively-geared property exposure with cheaper debt, and you don't sign a personal guarantee.

BTL landlords almost never include the value of their own time in the return calculation. A typical small portfolio (one to three properties) absorbs 50-100 hours a year of admin: tenant viewings, gas certificates, contractor scheduling, reference checking, end-of-tenancy disputes, MTD-for-landlord tax filing under HMRC's Making Tax Digital rules from April 2026. At even £30 an hour of opportunity cost, that's £1,500-£3,000 a year of unpriced work — and it's the unglamorous kind nobody actually enjoys.

REITs require zero hours a year. The professional management fee is already inside the share price. The diversification across hundreds of tenants is already done. The income tax wrapper is already arranged. You buy the share, you collect the dividend, you sleep.

For the rare landlord who genuinely enjoys property management and treats it as a vocation, the time isn't a cost — it's a hobby that pays. For everyone else, including most accidental landlords who inherited a parent's flat, the opportunity cost is real and rarely faced honestly. The BTL return brochure that doesn't subtract the landlord's own labour is a brochure designed to mislead.

Three groups should still consider direct buy-to-let in 2026: career landlords running incorporated portfolios who can use limited-company tax structures to dodge Section 24, professional renovators who add genuine value through refurb (the BRRRR strategy), and family vehicles where below-market rent to a child is part of a wider plan. None of those is the typical reader of this article.

For everyone else — the higher-rate-PAYE saver with £100,000 to deploy and no specialist landlord skills — the REIT route in an ISA dominates on after-tax return, on liquidity, on diversification, on time, and on regulatory exposure. The buy-to-let argument that the partner piece in this debate makes is the strongest version of the leverage case, but even there the leverage maths now require capital growth assumptions that recent ONS data does not support.

If you're already a landlord, the question is whether the after-tax yield on your existing properties beats what a like-for-like REIT exposure would deliver inside an ISA. If it doesn't — and for most higher-rate landlords it doesn't — you have a reallocation question, not a do-nothing question. For background context, see our property investing guide, the BTL tax-trap walkthrough, and the 2026 BTL tax-changes summary. The honest comparison wins this debate decisively for REITs at every standard portfolio size.

This article is for informational purposes only and does not constitute financial advice. You should seek independent financial advice before making any investment decisions. Capital is at risk. The value of investments can fall as well as rise. Past performance is not a reliable guide to future returns. Tax rules can change and depend on individual circumstances. For free, impartial guidance, see MoneyHelper.