Hong Kong's CCL Index Is Up 12% in 2026 — Sha Tin First City Still Sold at a HK$1M Loss

In one line: Hong Kong’s Centa-City Leading Index (CCL) hit 162.02 points in September 2026, up 12.43% year-to-date and its second-highest level in roughly three years — but in that same window, individual units at Sha Tin First City and Tai Po Grand Deer both resold at a loss. The CCL is a weighted average across about 100 large estates, refreshed once a week; it can’t tell you what your target estate’s asking prices and deals look like right now. To see that, you need to pull the estate-level data yourself.

A rising city-wide index doesn’t mean your estate is rising

According to Centaline’s own English data page, the Centa-City Leading Index (CCL) was last updated on September 4, 2026 at 162.02 points, up 0.30% week-on-week and 1.38% month-on-month. Hong Kong financial outlet etnet reported the same release in more detail: a 2026 year-to-date gain of 12.43%, and the index back above 162 points — its second-highest reading in the 157 weeks (about three years) since late August 2023. The four district sub-indices moved in different directions in that same week — New Territories West up 2.38%, Hong Kong Island up 0.31%, Kowloon down 0.38%, New Territories East down 1.04% — which is a first hint that “Hong Kong property prices are up” is not one uniform story even at the district level, let alone the estate level.

Individual transaction records make that gap concrete. A South China property report from Hong Kong Economic Journal documents a 2-bedroom, 327-sq-ft unit at Sha Tin First City, Block 19, low floor, Unit C: the owner bought it in August 2020 for HK$5.55 million and sold it in March 2026 for HK$4.55 million — a HK$1 million paper loss, about 18%, after a 6-year hold, at a saleable-area unit price of HK$13,914/sq ft. A separate report from HK01 records a 3-bedroom, 757-sq-ft unit at Tai Po Grand Deer, Block 1, high floor, Unit A: bought for HK$11.025 million in 2020, sold for HK$9.128 million in February 2025 — a loss of about HK$1.897 million, roughly 17.2%, after a 5-year hold. Both sales closed while the CCL headline was trending up.

Why the mismatch? Per the City University of Hong Kong Business School’s own description of the index, the CCL is built jointly by Centaline Property Agency and CityU’s Department of Decision Analytics and Operations, using a regression model that adjusts transaction prices for floor area, building age, location, direction, view and floor level — then rolls that up into one citywide number (plus district and unit-size sub-indices), published weekly every Friday after 4pm, on a base of 100 set in July 1997. Its underlying sample draws from roughly 100 large Hong Kong housing estates. That means: if your target estate is one of the ~100 large samples, its individual swings still get averaged in with everyone else’s and smoothed by a weekly cadence; if it’s a smaller estate outside that sample — Tai Po Grand Deer, for instance — the CCL was never tracking it in the first place.

Official index vs. raw estate-level data — what each one actually shows

How to track your target estate yourself, instead of relying on one weekly headline

Step 1 — Pull the CCL headline and its 7 sub-indices for market context

{ "mode": "ccl", "weeks": 52 }

weeks controls how many weeks of the citywide series: "ccl" you get back; leave it blank for the default 52 weeks, or set 0 for the full history back to 1994. The same run also returns the latest reading for each of the 7 sub-indices — Hong Kong Island, Kowloon, New Territories East, New Territories West, large units, mass units, and large estates — each with its own change_pct_wow and publish_date. This step tells you the macro backdrop, not what your specific estate is doing.

Step 2 — Pull the actual historical deals for your target estate

{ "mode": "transactions", "keywords": ["Sha Tin First City"], "dealType": "buy", "dateFrom": "30d", "maxItems": 50 }

keywords requires the estate name (one estate per keyword works best). dateFrom accepts a relative window (e.g. 30d for the last 30 days) or an absolute date; run this on a schedule and each pass only returns new deals, so your dataset accumulates into that estate’s own price timeline. Each record includes transaction_date, price and net_unit_price — you’re not waiting for Friday’s CCL release to see it.

Step 3 — Compare that against current asking prices

{ "mode": "search", "keywords": ["Sha Tin First City"], "dealType": "buy", "source": "both", "maxItems": 50 }

search mode returns what’s currently listed for that estate (price, price_per_sqft), which you can hold up against Step 2’s actual closed unit price (net_unit_price) to see how far the asking price has drifted from what’s really trading. To cover more agencies than Centaline and 28Hse, midland-scraper and ricacorp-scraper support the same search mode; 28Hse’s equivalent is mode: "listing". midland-scraper also has its own transactions mode, but only Centaline publishes the CCL index, so mode: "ccl" exists on centaline-scraper alone.

What this can’t do — it fills in the estate-level gap that the official index doesn’t show you; it isn’t a forecasting model, and it can’t tell you an individual seller’s actual bottom line or where prices go next. The buying or selling decision is still yours to make.

What it actually costs

All three modes bill under the same search-listing PPE event, priced per item returned:

A single run combining “CCL headline (52 weeks + 7 sub-indices, ~59 rows) + 50 transaction records for your target estate + 50 current listings” costs roughly $0.32 total — a lot closer to reality than trying to judge your target estate off a single weekly city-wide average.

Three ways to start

From least effort to most freedom

I run it for you Managed, daily to your sheet You do nothing. From US$299/mo. Use it here Run the free tool Instant, no setup, on this site. Run my actor On Apify, your account Same data, you run it and pay per result. Build your own Sign up for Apify Free account — build the pipeline yourself.

This post contains Apify affiliate links (no extra cost to you). Costs and comparisons are stated honestly, including where Apify is not the best fit.

FAQ

How much has Hong Kong’s CCL index risen in 2026? According to Centaline’s own English-language data page, the Centa-City Leading Index (CCL) stood at 162.02 points as of September 4, 2026, up 0.30% week-on-week and 1.38% month-on-month. etnet reported the same reading as a 12.43% year-to-date gain and the index’s second-highest level in the 157 weeks (roughly three years) since late August 2023.

Who compiles the CCL, and how many estates does it cover? Per City University of Hong Kong’s Business School, the CCL is compiled jointly by Centaline Property Agency and CityU’s Department of Decision Analytics and Operations, using a regression model that adjusts transaction prices for floor area, age, location, direction, view and floor level. It’s published every Friday after 4pm, with a base value of 100 set in July 1997. Its underlying sample draws from roughly 100 large Hong Kong housing estates, aggregated into one citywide figure plus district and unit-size sub-indices.

If the CCL is rising, does that mean the estate I’m watching is rising too? Not necessarily. A 2-bedroom unit at Sha Tin First City sold for HK$4.55 million in March 2026, versus the HK$5.55 million the owner paid in August 2020 — a HK$1 million paper loss, about 18%, after a 6-year hold. A 3-bedroom unit at Tai Po Grand Deer sold for HK$9.128 million in February 2025, against a HK$11.025 million purchase in 2020 — a roughly HK$1.9 million loss, about 17.2%. Both sales happened while the CCL headline was trending upward overall.

What’s the difference between centaline-scraper’s ccl mode and transactions mode? ccl mode returns the official CCL headline series plus its 7 sub-indices on a weekly cadence, controlled by a weeks parameter (defaults to 52 weeks if left blank). transactions mode requires a keywords value naming a specific estate, and returns that estate’s actual historical deals, including price, transaction date and unit price. The two modes operate at different levels of the market and can’t substitute for each other.

How is this actor priced? The search, transactions and ccl modes all bill under the search-listing event at $0.002 per item returned. The detail mode, which returns a full property record, bills under the product-detail event at $0.008 per item.

Chad runs 40+ published Apify actors, including the centaline-scraper used here. The CCL figures above are cross-checked against Centaline’s own English data page and etnet’s report on the same release; the two loss-making transactions are cross-checked against Hong Kong Economic Journal and HK01’s original reporting; pricing figures are cross-checked against the actor’s Store page and Apify’s official PPE documentation. Claims that couldn’t be sourced aren’t in this article.