${title}

29:null 2a:["$","title",null,{"children":"Did pay growth repair a decade of housing affordability? | Matt Lenzie | Observable"}] 2f:T774,# Did pay growth repair a decade of housing affordability? ## England and its nine regions, 2015–2025 Matt Lenzie, Construction Capital | Research note B001-R25 | 15 September 2026 England's median house-price-to-earnings ratio fell from 9.06 in 2021 to 7.63 in 2025. Look back to 2015, however, and the starting point was 7.54. The recent improvement had brought the national ratio close to its level a decade earlier. The regional picture is less uniform. All nine English regions had lower median ratios in 2025 than in 2021, yet seven remained above their 2015 levels. London and the North East were the exceptions. The comparison changes with the starting year, which is why this explorer lets you choose both endpoints. This Construction Capital analysis separates the arithmetic effects of house prices and earnings on the ratio. It uses the ONS workbook released on 26 March 2026, rechecked on 15 September 2026, with 110 observations and 550 possible area-period comparisons. The calculation adds a different lens to our earlier [affordability report for England and Wales](https://www.docdroid.com/W8vW51a/housing-affordability-improved-but-unevenly-england-and-wales-2021-2025-pdf). ## What the explorer shows Start with England and the full decade. Then compare a region or choose 2021–2025 to see the recent recovery. Orange bars show the price contribution to the change in the ratio; teal bars show the earnings contribution. A black point marks their sum. A negative net change means a lower price-to-earnings ratio. Contributions are measured in ratio points, not percentage points or pounds. The line chart puts the selected area's house prices and earnings on the same starting index of 100. It shows which grew faster without putting pounds of house value and annual earnings on one scale. The table supplies the original ratio levels and changes for England and every region.30:T28ba,// MIT licence. Construction Capital, 2026. Data: ONS under OGL v3.0. // Observable notebook cell: attach observations.json before running. const affordabilityData = await FileAttachment('observations.json').json(); function createAffordabilityExplorer(data) { const geos = [...new Set(data.map(d => d.geography))]; const lookup = new Map(data.map(d => [d.geography + ':' + d.year, d])); const root = document.createElement('section'); root.className = 'cc-affordability'; root.innerHTML = `

Start year End year Area to highlight

Price and earnings contributions to the change in the median ratio. Negative net change means a lower ratio.

PricesEarningsNet change

House pricesEarnings

Compare the underlying numbers

Swipe the table horizontally on a narrow screen. Ratio levels use ONS's published rounding; contributions use the component values.

Source: ONS, 26 March 2026 release, tables 1a–1c. Construction Capital calculations. These are arithmetic contributions, not causal estimates or mortgage affordability measures.

`; const $ = s => root.querySelector(s); const start = $('.start'), end = $('.end'), geo = $('.geo'); const option = (s,v,t=v) => s.add(new Option(String(t),String(v))); for(let y=2015;y<2025;y++) option(start,y); for(let y=2016;y<=2025;y++) option(end,y); geos.forEach(g=>option(geo,g)); end.value='2025'; const f = (n,d=2) => Number(n).toFixed(d); const sign = n => (n>0?'+':'')+f(n); let current=[]; function comparison(g,s,e) { const a=lookup.get(g+':'+s),b=lookup.get(g+':'+e); const p=(b.price_gbp-a.price_gbp)*(1/a.earnings_gbp+1/b.earnings_gbp)/2; const pay=(a.price_gbp+b.price_gbp)*(1/b.earnings_gbp-1/a.earnings_gbp)/2; return {geography:g,start_year:s,end_year:e,ratio_start:a.ratio_published,ratio_end:b.ratio_published, price_change_pct:100*(b.price_gbp/a.price_gbp-1),earnings_change_pct:100*(b.earnings_gbp/a.earnings_gbp-1), price_contribution:p,earnings_contribution:pay,net_change:b.price_gbp/b.earnings_gbp-a.price_gbp/a.earnings_gbp}; } const svg = (w,h,title,content) => ``; function render() { const s=+start.value,e=+end.value,g=geo.value; [...end.options].forEach(o=>o.disabled=+o.value<=s); [...start.options].forEach(o=>o.disabled=+o.value>=e); current=geos.map(x=>comparison(x,s,e)); const chosen=current.find(x=>x.geography===g); $('.summary').textContent=`${g}, ${s}–${e}: the ratio moved from ${f(chosen.ratio_start)} to ${f(chosen.ratio_end)}. Price contribution ${sign(chosen.price_contribution)} points; earnings contribution ${sign(chosen.earnings_contribution)} points; net ${sign(chosen.net_change)} points.`; $('.bar-title').textContent=`Two components, one ratio: ${s}–${e}`; const w=Math.max(340,Math.min(940,root.clientWidth||820)),left=w<500?123:174,right=28; const lim=Math.max(0.5,Math.ceil(Math.max(...current.flatMap(d=>[Math.abs(d.price_contribution),Math.abs(d.earnings_contribution),Math.abs(d.net_change)]))*2)/2); const x=v=>left+(v+lim)/(2*lim)*(w-left-right), h=440; let b=''; [-lim,-lim/2,0,lim/2,lim].forEach(v=>{b+=`${f(v,1)}`;}); current.forEach((d,i)=>{ const y=34+i*38; if(d.geography===g)b+=``; const name=d.geography==='Yorkshire and The Humber'?'Yorkshire & Humber':d.geography; b+=w<500&&d.geography==='Yorkshire and The Humber'?`Yorkshire &Humber`:`${name}`; [[d.price_contribution,-11,'#c56b28','Prices'],[d.earnings_contribution,2,'#167c7e','Earnings']].forEach(([v,offset,col,label])=>{b+=`${d.geography}: ${label} ${sign(v)} ratio points`;}); b+=`${d.geography}: net ${sign(d.net_change)} points`; }); $('.bar').innerHTML=svg(w,h,`Price and earnings contributions, ${s} to ${e}, ratio points`,b); $('.line-title').textContent=`${g}: prices and pay from the same starting point`; $('.line-subtitle').textContent=`Nominal values indexed to ${s} = 100. Each line uses its own starting value. This is not an inflation-adjusted house price index.`; const a=lookup.get(g+':'+s), series=data.filter(d=>d.geography===g&&d.year>=s&&d.year<=e); const points=series.flatMap(d=>[100*d.price_gbp/a.price_gbp,100*d.earnings_gbp/a.earnings_gbp]); const lo=Math.floor(Math.min(...points)/10)*10-5,hi=Math.ceil(Math.max(...points)/10)*10+5; const lx=y=>42+(y-s)/(e-s)*(w-65), ly=v=>230-(v-lo)/(hi-lo)*200; let l=''; for(let t=Math.ceil(lo/10)*10;t<=hi;t+=10)l+=`${t}`; const ticks=[...new Set([s,Math.round(s+(e-s)/2),e])]; ticks.forEach(y=>l+=`${y}`); [['price_gbp','#c56b28','House prices'],['earnings_gbp','#167c7e','Earnings']].forEach(([key,col,label])=>{ l+=``; series.forEach(d=>l+=`${d.year} ${label}: index ${f(100*d[key]/a[key],1)}`); }); $('.line').innerHTML=svg(w,275,`${g}: house prices and earnings, ${s} equals 100`,l); $('.table-wrap').innerHTML=`${current.map(d=>``).join('')}

Median affordability comparison, ${s}–${e}
Area	Ratio ${s}	Ratio ${e}	Price change %	Pay change %	Price effect	Pay effect	Net points
${d.geography}	${f(d.ratio_start)}	${f(d.ratio_end)}	${f(d.price_change_pct,1)}	${f(d.earnings_change_pct,1)}	${sign(d.price_contribution)}	${sign(d.earnings_contribution)}	${sign(d.net_change)}

`; } [start,end,geo].forEach(el=>el.addEventListener('change',render)); $('button').addEventListener('click',()=>{ const keys=Object.keys(current[0]),csv=[keys.join(','),...current.map(r=>keys.map(k=>typeof r[k]==='number'?r[k]:'"'+String(r[k]).replaceAll('"','""')+'"').join(','))].join('\n'); const url=URL.createObjectURL(new Blob([csv],{type:'text/csv;charset=utf-8'})),a=document.createElement('a'); a.href=url;a.download=`construction-capital-affordability-${start.value}-${end.value}.csv`;a.click();setTimeout(()=>URL.revokeObjectURL(url),1000); }); render(); const observer=new ResizeObserver(render);observer.observe(root); return root; } display(createAffordabilityExplorer(affordabilityData)); 31:T17d3,## Pay growth nearly offset the national price rise Between 2015 and 2025, England's median transaction price rose from £209,950 to £300,000, an increase of 42.9%. Median gross annual workplace earnings rose from £27,841 to £39,298, or 41.2%. In the symmetric calculation used here, the price change adds 2.76 ratio points and the earnings change subtracts 2.67. Together they produce a rise of about 0.09 points. These are arithmetic contributions: the calculation does not estimate what caused prices or earnings to change. The post-2021 comparison is stronger. Prices rose 5.3%, while earnings rose 25.0%. Their contributions are approximately +0.43 and −1.86 ratio points, giving a net decline of 1.43. The national ratio improved even though the median transaction price increased. ## A recent recovery can coexist with a longer squeeze The East Midlands had the largest increase in the median ratio across the nine regions over 2015–2025: from 6.28 to 7.02. Prices rose 59.2%, compared with earnings growth of 42.4%. Its price contribution of about +3.17 points exceeded the earnings offset of −2.42. London's ratio fell from 11.05 to 10.61 over the same decade, but it remained the highest of the nine regions in 2025. The North East also improved, from 5.15 to 5.00. Direction and level answer different questions: an improving area can still have a high ratio. These results suggest a useful discipline when reading a housing-market headline: check the endpoint, the starting year and the price and earnings components. An improvement since a recent peak does not establish that the earlier gap has closed. ## How this connects with the wider housing evidence The [earlier affordability report](https://www.docdroid.com/W8vW51a/housing-affordability-improved-but-unevenly-england-and-wales-2021-2025-pdf) covers median and lower-quartile ratios and the breadth of recent local-authority changes. Its [original backing package](https://drive.google.com/file/d/1a5mQc96GipZVnRrakIu1wA-gxYQ4WV5d/view?usp=drivesdk) preserves the wider country and local-authority data. Housing-stock growth answers a separate supply question. Our [2015–2025 housing-stock explorer](https://public.flourish.studio/visualisation/30240938/) and its [source data and code](https://drive.google.com/file/d/12jeKowZRze5gIDNEPo4BjiAmoPVKhQY9/view) track England and the regions by tenure. We have not used those stock figures to claim that supply caused the price-to-earnings changes shown here. The [Construction Capital market-reports collection](https://constructioncapital.co.uk/market-reports) brings together related work on housing supply, planning and property markets. These companion studies are also produced by Construction Capital; the hosting platforms do not provide independent endorsement. ## Method and interpretation Let P be the median house price, E median earnings and R = P/E. Subscripts 0 and 1 identify the selected start and end years. The calculation averages two paths: changing prices first and earnings second, then changing earnings first and prices second. This assigns their joint interaction equally. Price contribution = (P1 − P0) × (1/E0 + 1/E1) / 2. Earnings contribution = (P0 + P1) × (1/E1 − 1/E0) / 2. The contributions sum exactly to P1/E1 − P0/E0 before display rounding. They describe the selected endpoints, not the sum of annual contributions along the intervening path. We calculate from the published component values and retain ONS's rounded ratio levels for comparison. Tiny displayed differences can arise from rounding. England's national figures are used directly, rather than averaged from regional medians. Prices cover transactions in the year ending September; earnings use the April ASHE measure for gross full-time employees working in the area. The 2025 earnings estimates are provisional. Transaction medians can change with the mix of homes sold. Earnings estimates carry sampling uncertainty, and pandemic effects and methodological changes affect comparisons. No new significance tests are made. The two medians do not match individual buyers to their earnings. The ratio excludes deposits, borrowing costs, joint household income and lending criteria. It cannot establish a household's ability to purchase, or a development's viability. See the [ONS quality and methodology information](https://www.ons.gov.uk/peoplepopulationandcommunity/housing/methodologies/housingaffordabilityinenglandandwalesqmi) for source definitions and limitations. ## Data, code and reuse [Download the complete backing data and reproduction code](https://drive.google.com/file/d/1LKHtqE8icX7TLa2wYuQoJS7EveXOr8VD/view). The same 22-file archive is attached to this notebook. It includes the original ONS workbook, four CSVs, source observations, methods, validation checks and code. The accompanying archive contains the preserved ONS workbook, source-cell references, the full comparison table, validation checks, reproduction code and the notebook text. Four CSV outputs and both JSON outputs reproduce from the preserved workbook. The interactive display recalculates its selected comparison from those source observations. Source: [ONS, House price to workplace-based earnings ratio](https://www.ons.gov.uk/peoplepopulationandcommunity/housing/datasets/ratioofhousepricetoworkplacebasedearningslowerquartileandmedian), tables 1a–1c, released 26 March 2026. Contains public sector information licensed under the [Open Government Licence v3.0](https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/). Construction Capital's analysis and text are available under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/); code under the MIT licence. Attribute the original ONS source and Construction Capital's calculations separately. Construction Capital arranges property finance and has a commercial interest in the housing market. This is a descriptive research note, not a peer-reviewed study or a lending recommendation. Version 1.0. Corrections can be sent to matt.lenzie@construction-capital.co.uk. 32:[{"id":"08eb5ad5e79c1c619e791fc3371bd320da81f0c13b994009f44ad546b393d41cd339f3c4d128394eb9f8f10aa364752828998f8dea931e43bde9dd1d8a3e41b3","href":"https://static.observableusercontent.com/files/08eb5ad5e79c1c619e791fc3371bd320da81f0c13b994009f44ad546b393d41cd339f3c4d128394eb9f8f10aa364752828998f8dea931e43bde9dd1d8a3e41b3","name":"observations.json","type":"application/json","lastModified":1789461971585,"size":54991},{"id":"96dcf9442984ccb0cc4d1c413c4cb9b597279741409ee9659b2c26a4a7d67585fce7b8e8001946e6a05f05cf774b1a8780d97d5d718f49e46474b614bd7abe2e","href":"https://static.observableusercontent.com/files/96dcf9442984ccb0cc4d1c413c4cb9b597279741409ee9659b2c26a4a7d67585fce7b8e8001946e6a05f05cf774b1a8780d97d5d718f49e46474b614bd7abe2e","name":"B001-R25-affordability-decomposition-backing-data.zip","type":"application/zip","lastModified":1789462353143,"size":586515}] 2d:{"id":"37e2828f32d32467","slug":"did-pay-growth-repair-a-decade-of-housing-affordability","title":"Did pay growth repair a decade of housing affordability?","thumbnail":"8694434e61154420ea4dd3278103a0fd5e9eacba652ddfa2b5a0a042c1ebfc3b","access_level":"public","create_time":"$D2026-09-16T15:08:41.983Z","update_time":"$D2026-09-16T15:18:55.105Z","view_time":null,"first_public_time":"$D2026-09-16T15:19:19.235Z","has_unpublished_edits":null,"fork":false,"stars":0,"starred":false,"owner":{"type":"team","id":"ea2de9ba2c0a2ed6","login":"matt-lenzie","name":"Matt Lenzie","avatar_url":null,"tier":"free_2026"},"version":12,"publish_version":12,"role":null,"authored":false,"authors":[{"type":"individual","id":"2a28c9706791cd4f","login":"matt-lenzie","name":"Matt Lenzie","avatar_url":null,"tier":null}],"members":[{"type":"team","id":"ea2de9ba2c0a2ed6","login":"matt-lenzie","name":"Matt Lenzie","avatar_url":null,"tier":"free_2026","role":"editor"}],"comments":[],"custom_thumbnail":false,"trashed":false,"lapsed":null,"subscribed":null,"forks":0,"collection":null,"type":"notebook","body":{"title":"Did pay growth repair a decade of housing affordability?","cells":[{"id":1,"value":"$2f","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":2,"value":"$30","pinned":false,"hidden":false,"mode":"ts","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":3,"value":"$31","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"}],"files":"$W32","stdlib":"2","titling":"auto"}} 2e:"$2d:body"