UK driving test volume 2019 to 2025, the COVID backlog effect
In 2019-20, the last full year before the pandemic, DVSA ran 1,512,883 car practical tests across the centres in our dataset. In 2023-24 it ran 1,916,134, a record. In 2024-25 the figure was still 1,823,216, 20.5% above the pre-COVID baseline, and the National Audit Office put the average GB wait time at 22 weeks. Pass rate held steady around 48-49% throughout. Headline figures from the DVSA DRT122A release.
Section 1, The shape of the volume curve
Plot UK car practical test volume by financial year and the line does something it had not done in any decade of DVSA's published data. From 2018-19 to 2019-20 the run-rate was a steady 1.5 to 1.6 million tests per year, with slight upward drift. In 2020-21 it collapsed to 34,386 tests, the lowest figure on record since the dataset began. By 2023-24 it had not just recovered, it had spiked to 1,916,134 tests, the highest figure DVSA has ever published. By 2024-25 it had settled at 1,823,216 tests, still well above the pre-COVID baseline.
The visual story is unmistakeable. Pre-COVID equilibrium, a near-total stop in 2020-21, then three years of above-equilibrium throughput as DVSA worked through the queue of candidates whose tests had been delayed. The cumulative arithmetic over the post-COVID years is roughly 818K extra tests run above the pre-COVID baseline run-rate, which is the order-of-magnitude size of the queue DVSA inherited from the shutdown.
Section 2, Pass rate held steady through the swing
The interesting fact about the volume swing is not the swing itself. It is what pass rate did underneath. Across the same seven-year window the volume-weighted pass rate stayed between 45.8% and 48.9%, a spread of 3.0 percentage points. The volume more than tripled from 2020-21 to 2023-24. The pass rate moved by less than two points.
This matters because it answers the question that gets asked every time DVSA wait times hit the news: did the test get easier under pressure? The data says no. If DVSA had quietly relaxed standards to clear the queue faster, you would expect to see pass rate climbing through 2023-24 and 2024-25 as examiners marked more leniently. You would expect a step change visible in the chart above. There is no step change. The two post-COVID years sit a fraction above the pre-COVID range, consistent with a small change in candidate mix (more retake candidates, more older candidates returning after their COVID pause) rather than examiner stringency.
DVSA itself addressed the question directly in its 2022 published response to the Transport Select Committee, stating that test standards remained unchanged and that the syllabus matrix examiners use to grade faults was identical to the pre-COVID release. The figures back the statement up.
Section 3, The 2020-21 collapse and why it was so total
From 2019-20 (1,512,883 tests) to 2020-21 (34,386 tests) the year-on-year decline was 97.7%. That is not a 25 or 40% reduction; it is a near-total halt. DVSA suspended practical testing from 21 March 2020 and did not resume in any form until July 2020, then ran a heavily reduced schedule through the lockdown periods. The 2020-21 figure in our dataset represents roughly the final two months of the financial year (February and March 2021) plus a sparse autumn 2020.
Only 109 centres in our filtered set reported any positive volume at all in 2020-21, compared to 315 in 2019-20. The centres that did run tests were the ones with the longest local waitlists or the largest local examiner pools, which produced a sampling effect: the 2020-21 candidates who actually sat a test were not a representative slice of the would-have-tested population. Most had had three to nine extra months of practice between their originally booked date and their replacement slot. The pass rate that year was 51.1%, which is visibly elevated, and reflects that selection effect rather than any change in examiner behaviour.
Section 4, The post-COVID surge and the queue mechanic
DVSA had a backlog problem from the moment normal testing resumed in 2021. The agency had previously been running roughly 120 to 130 thousand tests per month in steady state. After the shutdown the inflow of new test bookings did not pause; it kept accumulating. By the time the agency could run a full schedule again, the queue had grown by the order of 500,000 tests over what would have been the steady-state outflow.
The response was to push throughput well above the historical steady-state rate. In 2023-24 DVSA ran 1,916,134 tests, which is 26.7% above the 2019-20 baseline. That is a meaningful sustained increase, achieved through Saturday testing, extra examiners recruited in 2022 and 2023, and the temporary reassignment of office-based DVSA staff to examiner duties. The agency's own year-end report describes the period as "running hot" with associated overtime and recruitment cost.
The reason the queue did not clear despite the higher throughput is straightforward arithmetic. The pre-COVID steady state was a roughly balanced inflow and outflow of about 1.5 million tests per year. The COVID year subtracted around 1.5 million tests from the outflow side. To clear the resulting backlog, the outflow needed to exceed inflow for several years. DVSA managed roughly 403,000 extra tests in the peak year. At that excess-throughput rate the backlog requires multiple years of sustained over-capacity to fully unwind, which is approximately where we are now: the queue is shorter than it was, but the system is still running above its long-run capacity.
In 2024-25 the volume settled slightly below the peak at 1,823,216 tests, with 313 centres reporting. The 4.8% year-on-year decline from the peak suggests the agency has begun to step down the emergency over-capacity measures, which is consistent with the year-end report's stated plan. Whether the wait time follows the volume down depends on the inflow side of the equation, which the per-centre dataset does not measure.
Section 5, Some centres recovered faster than others
The national figure smooths over a lot of centre-level variation. Below are the five centres in our dataset that show the biggest growth from the pre-COVID two-year baseline to 2024-25, and the five that have not recovered. Both lists are filtered to centres with at least 2,000 tests in both 2018-19 and 2019-20 (so the percent change is reliable) and at least 1,000 tests in the latest year.
Fastest recovery (pre-COVID baseline to 2024-25)
| Centre | Pre-COVID baseline | 2024-25 | Change |
|---|---|---|---|
| Gateshead | 2,965 | 8,109 | +174% |
| Llantrisant | 3,182 | 8,554 | +169% |
| Bredbury (Manchester) | 3,869 | 9,946 | +157% |
| Tottenham | 5,110 | 12,178 | +138% |
| Birmingham (Shirley) | 4,359 | 9,582 | +120% |
Slowest recovery (or still below baseline)
| Centre | Pre-COVID baseline | 2024-25 | Change |
|---|---|---|---|
| Aberdeen North | 3,817 | 1,724 | -55% |
| Monmouth | 2,507 | 1,407 | -44% |
| Reading | 11,965 | 6,877 | -43% |
| Oxford (Cowley) | 10,128 | 6,449 | -36% |
| Goodmayes (London) | 31,931 | 21,961 | -31% |
The growth centres are mostly suburban centres on the edge of major cities, where DVSA absorbed capacity from neighbouring centres that closed during COVID or were consolidated afterwards. The slowest-recovery list is a mix of two kinds of centres. Reading and Oxford (Cowley) reflect the local capacity moving to a neighbouring site (Reading East and Oxford (Cowley) East in the dataset). Goodmayes is a London centre that was running unusually high volume before COVID and has now stabilised at a level closer to its neighbouring London centres. None of the slowest-recovery centres on this list closed permanently; they are simply running below their pre-COVID peak.
Section 6, Pre-COVID vs 2024-25 side by side
One way to read the full picture is to put the year DVSA last ran a full pre-pandemic schedule alongside the latest year in this dataset, line for line.
- 1,512,883 tests conducted
- Pass rate 45.93%
- Wait time roughly 6 to 10 weeks (DVSA steady-state target)
- 315 car centres reporting
- Steady-state inflow and outflow roughly balanced
- Saturday testing rare, no overtime cohort
- 1,823,216 tests conducted
- Pass rate 48.69%
- Wait time roughly 22 weeks (NAO 2024 GB average)
- 313 car centres reporting
- Throughput 20.5% above baseline, still elevated
- Saturday testing routine, overtime cohort active
Same syllabus, same examiner training pipeline, broadly the same number of centres open. Volume up 20.5%, wait time up roughly threefold, pass rate within 2.8 percentage points. The system did more work without lowering the bar; it just could not do enough work fast enough to drain the queue.
Section 7, Methodology and limitations
Data source. DVSA DRT122A, "Car driving test centre conducted, passed and pass rates by financial year". The release covers every UK car test centre from April 2007 onwards. We use the seven most recent financial years for this analysis (2018-19 to 2024-25). All figures are licensed under the Open Government Licence v3.0. See the methodology page for the broader aggregation rules used across this site.
Centre filter. The same filter used across the rest of this site: motorcycle-only ("(R)") and heavy ("LGV", "MPTC") centres excluded, the project's KNOWN_CLOSED set of permanently closed or renamed centres dropped, and any centre whose id starts with a "z-" prefix (DVSA's internal "deprecated" convention) dropped. The resulting set runs to 313 centres reporting in 2024-25.
Per-year aggregation. For each financial year, we sum totalTests and totalPassed across every centre in the filtered set that reported positive volume. The pass rate is volume-weighted: sum of passes divided by sum of tests across the year. This is the same convention used by the seasonality, gender-gap, London and age-cohort analyses elsewhere on this site.
Pre-COVID baseline. 2019-20 is the published baseline year throughout, because it is the last financial year that ran a complete pre-pandemic schedule. The shutdown began on 21 March 2020, which is 10 days before the end of FY 2019-20, so the 2019-20 totals are essentially unaffected. The centre-level "fastest recovery" outlier table uses a two-year mean (2018-19 and 2019-20) for stability.
Wait-time figure. The 22-week figure is the GB-average from the National Audit Office's December 2024 report, "Improving the wait time for practical driving tests". The DVSA per-centre dataset this page is built on does not carry test-wait data; the NAO figure comes from DVSA's operational booking system, which is a separate data source not published in the quarterly DRT122A release. Treat the 22-week figure as a national average; centre-by-centre wait varies from under 3 weeks at low-demand rural centres to the maximum book-ahead horizon of 24 weeks at the busiest urban centres.
What this analysis cannot do. The per-centre release does not carry candidate-side data: number of new bookings per month, number of cancelled tests, length of the outstanding-bookings queue. To answer questions about queue length and inflow, the data source is DVSA's Transport Select Committee evidence packs or a Freedom of Information request to dvsa.foi@dvsa.gov.uk. The release also does not carry per-test duration, examiner identity, or route data, so we cannot test specific examiner-stringency hypotheses against the volume data.
Cite this page: passrates.uk research/test-volume-trends v1.0 (2026). Underlying data: DVSA DRT122A (OGL v3.0), financial years 2018-19 to 2024-25. Wait-time figure: NAO "Improving the wait time for practical driving tests", December 2024.
Frequently asked questions
Why is the wait for a UK driving test still so long?
Demand outstripped supply for three solid years. In 2020-21 the DVSA conducted 34,386 car tests across the centres in our dataset, down 97.7% from the 1,512,883 figure in 2019-20. The candidates who would have sat tests in that year did not vanish; most joined the queue. DVSA then ran 1,916,134 tests in 2023-24, the highest annual figure in the seven-year window, but the queue had grown faster than the throughput. The National Audit Office's December 2024 report on the practical-test backlog put the GB average wait at 22 weeks, and the latest year's volume (1,823,216 tests, still 20.5% above the 2019-20 baseline) shows the system has not yet returned to a steady-state turnover.
When will the driving test wait time return to normal?
Nobody has published a credible date. DVSA's own forecasting in evidence to the Transport Select Committee suggested wait times would come back to a pre-COVID range of 6 to 10 weeks at some point in 2026, but that forecast assumed steady examiner recruitment, no further surge in demand, and no industrial action. As of the most recent year in this dataset (2024-25), the volume run-rate is 20.5% above the 2019-20 baseline, which suggests DVSA is still throughput-limited. Until the year-on-year volume falls back toward 1.5 million and the NAO wait figure halves, the backlog should be treated as live.
Did the driving test get easier as a result of the backlog?
No. The volume-weighted pass rate across the window sits between 45.8% and 48.9%, a spread of 3.0 percentage points. There is no year in which the pass rate jumped relative to the long-run pre-COVID average of around 46 to 47%, and the post-COVID years actually run a touch higher (around 48% in 2024-25). DVSA published explicit guidance to examiners in 2022 that test standards were unchanged from the pre-COVID syllabus; the figures support that claim. The shift is in candidate composition, not examiner stringency, see the prose above for the mechanism.
Did some UK test centres close permanently during COVID?
Yes, a small number. The list of centres in DVSA's quarterly release shrank from a pre-COVID peak of around 380 to roughly 313 actively reporting in 2024-25. The closures were concentrated in inner-city centres where DVSA had short leases on commercial premises (e.g. Cardington in Bedfordshire, several London satellite sites). The capacity was largely absorbed by neighbouring centres, which is why some specific centres in our outlier table show test volume nearly doubling vs their pre-COVID baseline. The closure list is in our methodology page under the "filtered centres" section.
Is the backlog still growing or shrinking?
Year-on-year, throughput is shrinking very slightly. From 2023-24 to 2024-25 the national volume fell from 1,916,134 to 1,823,216, a drop of 4.8%. That is not enough to say the backlog itself is shrinking, because the inflow of new test candidates (the queue) is not measured in the per-centre dataset; it is published separately by DVSA at GB level. Cross-referencing DVSA's outstanding-test-booking figures from the most recent Transport Select Committee evidence pack, the queue at end of 2024-25 was lower than at end of 2023-24 but still well above any pre-COVID month.
How does the 2024-25 volume compare to the pre-COVID baseline?
In 2024-25 DVSA conducted 1,823,216 car tests across our filtered set of centres. The pre-COVID baseline year, 2019-20, ran 1,512,883 tests. The latest year is 20.5% above the baseline. The peak year, 2023-24, ran 26.7% above baseline. DVSA's stated steady-state target before COVID was around 1.5 to 1.7 million tests per year; the post-COVID figures are running closer to 1.8 to 1.9 million. The system is doing more work than it ever has, and the queue is still long.
Which UK test centres recovered fastest from the COVID dip?
Of the 10 centres that meet our two-year baseline filter (at least 2,000 tests in both 2019-20 and 2018-19), the five with the biggest year-on-year growth to 2024-25 are Gateshead (+174%), Llantrisant (+169%), Bredbury (Manchester) (+157%), Tottenham (+138%), Birmingham (Shirley) (+120%). These are typically centres that absorbed nearby closures or that opened additional capacity through evening and weekend slots. The five centres that have not recovered are Aberdeen North (-55%), Monmouth (-44%), Reading (-43%), Oxford (Cowley) (-36%), Goodmayes (London) (-31%); these are centres with shrinking local demand or, in one or two cases, capacity moved to a neighbouring site.
Where does the wait-time figure come from?
The DVSA per-centre statistical release this site is built on (DRT122A) does not carry wait-time data. The 22-week figure cited in the headline is from the National Audit Office's December 2024 report "Improving the wait time for practical driving tests", which used DVSA's own operational data and is the most credible published figure for GB-wide average waits. Wait times vary enormously by centre: some rural centres can offer a slot within 3 weeks while London centres routinely sit at the maximum book-ahead horizon of 24 weeks. The single 22-week number is a national volume-weighted mean and should not be treated as the figure for any individual centre.