Morning.
Morning.
Thanks, Kevin. The chairman's the only one who responded. He's right in my aisle. Morning.
Morning.
Yeah, thank you. Right, see if we can get the clicker to work. I'm very conscious of time this morning because we've got three presenters and a reasonable amount to get through. My first section on overview and current trading, I want to get through fairly quickly so we don't waste too much time. Spend a little bit of time on current trading and, obviously, happy in questions to come back to that. The overview, hopefully, really will be an overview. This first slide sets out our key financial metrics. You've seen them before. There's no surprises in there. I think what I reflected on looking at it was going back to the start of the year and what we have expected. With one small change, it would've looked exceptionally similar to that.
That's true also for volume and the overall trading performance of the business. The one exception, you go back to the beginning of the year, and we'd have probably expected to have a lower cash conversion and a slightly lower return on net assets because we'd have expected to spend slightly more on land. I think with the sort of uncertainty during the course of the year, definitely as we came into the final quarter, we pulled back very slightly. I'm talking about, we're talking to the tune of GBP 100 million, GBP 150, something like that. The higher cash balance at the end of the year isn't just land. If I go back to the beginning of the year, that's the one thing that's slightly different. We've just been that edge more cautious as we've gone through the latter half of 2018.
This is a new set of KPIs, which you won't have seen before. I'm not, you'll be relieved to know, going to go through every one and define it for you and compare it for last year. Jennie will pick up the same slide at the beginning of her presentation. I did want to spend a minute or two on what the background to it is, though. We've gone through and looked at what measures we are using internally, and particularly on things like build quality and to a certain extent on customer satisfaction, although it's been there longer term. There are a new set of measures that we're using, both these and sort of things that tier out below them.
We wanted to give you as rounded a set of indicators that covered what we saw as the key strategic drivers of the business in line with the new strategy. It's deliberately new, but for instance, there is more on employee as well as on build quality. You can see on the employee numbers the big step up in particularly the number of new recruitments into early talent programs. You'd see the same on apprenticeships. Our apprenticeship numbers, new starters are about 50% higher in 2018 than 2017, and our expectation is there'll be another 50% higher in 2019 and probably higher again in 2020 as we roll out the pilots that we've done across all our businesses. Hopefully as we go through the next couple of years, this KPI sheet will give you some backdrop and a building picture of those underlying strategic measures.
Jennie, as I say, will pick up a couple of the specific ones, particularly around customer satisfaction and build quality. I'm leaving the customer satisfaction numbers entirely to her, even though that's 90%, which we're quite pleased with. Coming onto the market backdrop, I say 2018, but to be honest, I'll move quicker than I normally do into where we see things are today. The top block on that slide you've seen before. Just a sort of quick snapshot of where interest rates will be. I don't want you to worry too much about the 3.84% on Help to Buy equity loan. I think I'm right in saying that's Aldermore. I think the next highest is the Bank of Ireland, which is about 2.51%, then you drop down into exactly the same range as we had at July.
You still have, for our customers, a very low set of interest rates with quite a wide choice of lenders. Whilst not a lot is being said at the moment about mortgage lending from a sort of overall availability and cost point of view, I wouldn't want anybody to go away from this presentation and not realize how fundamentally important that is. I will come back at the end and talk about short-term, medium-term outlook. My own personal view is that in the end, this cycle will change because of interest rates, not because of Brexit or anything else. I think we see that probably being stable for a slightly longer period than people expect. That's what drives the underlying strength of our market one way or another. Those low interest rates are absolutely critical. The second block is new.
What we're trying to do, it is difficult, so I would ask you to take the signals of this in the round rather than looking at any one individual measure. We're trying to give you a sense of the leading indicators we look at market-wise and have been doing year-on-year. I'd pick out one of them in particular to talk about today, the second one on house price inflation. If I look back over the various housing markets I've seen over the last 20 years and go back to the U.S. and look back with hindsight, the signs that were building up as, say, before the market changed. It's actually that bubble effect of the significant price rises that happen late cycle that drive the triggers and the scale of a future downturn.
That sense that you look at the cumulative over three years of house price inflation is probably the best indicator of the level of market-related risk if the external environment remains the same, I think is a useful one. Is 15% cumulative the right level which to get worried? You can argue about it forever and a day about what the right level. Looking at that sort of measure is important. If I go back to some of those U.S. markets in Florida and Arizona, that measure would have told you 70% over the previous three years. You go back to the U.K. before the 2008, 2009 downturn, you'd end up at about 30%. 13.4% actually isn't particularly uncomfortable, particularly when you realize that nearly all of that was in the first of those three years.
In that time, it's dropped from about 6% to about 2.5%, and today is reasonably flat. I'm not trying to convince you that there isn't risk in this market. We'll come back to that a lot. I do think getting past the short term, we should be looking at some of the long-term underlying drivers of the market rather than short-term trading perhaps quite as much as we do. This slide you've seen before. I'm not going to go through every individual graph, but what you will hopefully see is that trading and our forward indicators we look at of our own trading, and we'll come back to sales rate in a second, remain in a pretty healthy place and remain very consistent with what we would expect at this time of year. No real warning signals there in any of those four indicators.
Probably more interestingly, and probably the sort of number that you would have picked out from your early read of the statement, if I look at trading in the first eight weeks of 2019, that private sales rate of 0.99, I have to say, is ahead of where we would have expected it to be. It's about where we wanted it to be, but it's ahead of where we would have expected. I'll come back to an individual larger scale sale in that, but even without that, the underlying sales rate will be 0.9, which is 10% ahead of where we were last year, in line with where we were the year before, which is an all time record. Very healthy. I would argue that tells you two things. One, the underlying market is okay and hasn't fundamentally changed in early 2019.
Two, some of the things we're doing on strategy on larger sites that we talked about last year, you can see in our late 2018 and early 2019 trading. I will be very surprised if that sales rate and the year-on-year performance and the order book doesn't lead the sector. We'd have been doing something wrong relative to our strategy if that wasn't the case. Hopefully you take a degree of comfort that we know what we're doing when you see those sorts of numbers. That leaves us with an order book which is significantly ahead year-on-year. You've got the numbers on the bottom bullet point. If I strip out the affordable element, which is where the larger percentage of the growth is, the private order book is still 7.5% ahead of this time last year.
Again, is ahead of where we would have expected. That doesn't particularly change our view of volume for this year, I'll come back to that at the end. Jennie.
Thanks, Pete. Good morning, everyone. Starting as always with the group results, 2018 was a record year for Taylor Wimpey. We generated more revenue in the U.K. than we ever have, even compared to the early days post-merger, when we also had a U.K. construction business. The group's gross profit and operating profit represent record performances. Both measures benefited from volume and margin growth and increased by just over 4%. The 30 basis point increase in the operating margin to 21.6% is particularly pleasing. Yes, that too is a record performance. PBT and adjusted EPS have increased by 5.5 and 5.4% respectively, which means we've been able to pay more to shareholders in dividends than ever before, and at the same time been able to invest in the business with the tangible net asset value per share increasing by 2.7%.
As we go through 2019, our ongoing investment in the business will continue to be disciplined as it has been in recent years in delivering that record cash performance in 2018, and a return on net operating assets at 33.4%. U.K. completions increased 3% in 2018, with all of that increase coming from affordable homes. The mix of affordable at just shy of 23% was greater than the affordable mix in the previous two years at 19%. Mix expectations going forward around about the 21% mark, assuming, of course, that there's no change in market conditions. This was mainly due to timing and reflected the delivery profile of the sites under development.
Year-on-year pricing showed individual improvement for both private and affordable homes, but the mix shift towards affordable meant that the average across the two was flat at GBP 264,000. The contribution from JVs reduced in 2018, I would expect that to be pretty flat in 2019 with the potential for growth from 2020 onwards as the Winstanley JV at Clapham starts to deliver. We started presenting this indicative analysis of margin movement back in 2014. This is the first period since then when the income statement impact of build cost inflation has exceeded that from selling prices, the net impact being 0.3%.
There was still a small positive pricing trend over the second half of 2018, but the rate of price inflation had been flattening out over the course of the year. That means we've got to work harder with costs in the future to manage that cost inflation and protect margin. Land bank evolution continues to show a small negative impact on the income statement, as the completion volumes from the older super margin sites continue to reduce, pardon me, as a portion of the total. That impact from land bank evolution is offset by a slightly greater impact from land mix, where better quality locations, design, and delivery have enhanced that financial outturn over and above the market performance. We previously reported enhanced customer journey as margin dilutive. That's as we increased our investment in customer service and build quality.
We're now seeing the benefit of that investment in a reduction in both current and longer term remediation costs. There was also a small benefit to margin in the period generated from land sales and some commercial property sales from our mixed-use developments in London. We're making good progress with the cost and efficiency program. There are numerous work streams running, and the ones on the slide are really just a selection to give you a feel for progress and what's being delivered. That first phase of delivery excellence is now fully deployed. This involved over 800 of our site staff getting new mobile devices, which have apps that are integrated into our ERP system, and that's going to increase the efficiency on site in lots of different ways.
For me, it's about allowing those site teams to spend less time in their office and more time out and about on site, ensuring that right first time build quality. Phase 2 of delivery excellence involves the deployment of even more of those apps, to support program delivery and better cost control. They're being tested in Q2 with delivery anticipated for Q3. We're expecting there'll be even more time saved from Phase 2 than from Phase 1. Commercial excellence is probably the biggest and most complex element of the overall program. It's tantalizingly close to deployment. Testing's going well. To put some color on that, we have already run live pilots on most of the work stream, so we're confident of being able to get underway with the rollout in Q2.
As I mentioned, I think back in July, we're expecting the first phase of commercial excellence to drive a lot of efficiency. Most of the 10,000 days saved will be redirected into activities that are more value-adding, and the delivery team are very focused on that as part of the rollout. One of them is in the audience today, so it's good for him to hear that. Some of that time saved in commercial excellence will be refocused onto groundworks. To ensure we get the best value from that redeployment, I'm currently running a separate project to develop groundworks trading for our commercial teams, to really try and bring some consistency to how we approach what can be a complex area of our business.
Despite the increase in the proportion of affordable homes delivered in 2018, we were able to achieve our objective of building more homes from our standard house type range, increasing that number from 65 to 78%. Since May last year, we've also been plotting standard house types from the consolidated range, which will bring further build and procurement benefits. In procurement, we now have a dedicated central team bringing more consistency and best practice to the management of our relationships with our national suppliers. Overall, it's still early days for the cost and efficiency program, but our focus remains on driving productivity through technology, standardization, and best practice. The combined private and affordable gross profit per unit decreased by GBP 400 in 2018 due to the greater proportion of affordable homes.
In both the table and the chart, you can see a switch between land and build cost per unit. The switch is partly driven by the greater proportion of affordable units, which have a lower land cost, pardon me, but also by a greater proportion of completions from large sites, which have a greater infrastructure demand and therefore lower land values as a result. There are two items reported as exceptional in 2018. The ACM cladding provision, you will remember from the half year reporting. Our estimate of these costs remains at GBP 30 million. Although that spend against the provision in 2018 at GBP 400,000 was relatively modest, there has been a significant amount of time and effort invested in designing the replacement solutions and preparing for those works to commence.
The second item is a charge relating to the guaranteed minimum pensions equalization for our defined benefit pension scheme, and that impacts on all defined benefit pension schemes in the U.K. We first reported this in the January trading statement, where we provided a range of GBP 15 million to GBP 20 million, and that final charge of GBP 16.1 is within that range. The strength of our balance sheet puts us in a really great position for the future. Land net of land creditors at just over two billion is 1% less on year-on-year, as we took advantage of attractive opportunities to negotiate deferred terms on larger sites. Our adjusted gearing, including land creditors, has actually reduced to 2.9% off the back of very strong cash generation. I'll come back to land creditors in a minute.
The land cost as a percentage of average selling price in the short-term owned land bank remains low at 15.2%, indicating the quality of that land bank and its ability to generate cash and profit in the future. WIP increased by 3%, which was in line with the increased volumes, as you might expect. Net assets also increased by 3%, even after paying dividends of GBP 500 million during the year. I think I've probably covered most of the points on this slide, but it's worth noting that last month we extended our GBP 550 million revolving credit facility by an additional year. It now expires in February 2024. That together with our EUR 100 million private placement loan, which expires in 2023, that brings the current weighted average life of our committed facilities to five years.
That gives us the certainty and the flexibility to take advantage of any opportunities that might present themselves. Jennie will talk a bit more about our approach to land in a minute. Given the increase in land creditors, I thought it might be interesting to look at them from a different angle. This slide shows you the distribution of the value of land creditors based on where those individual sites sit in the land quality matrix. 2008 and 2009 were pretty painful in lots of ways, but one of the things that sticks in the mind is paying land creditors on sites that you really wish you hadn't bought. Interestingly, the sites that make up the 11% in the amber cells, so the 5% and 1% cells.
On average in 2018, those sites delivered a sales rate in excess of one a week and a contribution margin of 27%, and included some pretty good locations in London, Surrey, and Buckinghamshire. I think that's a great indicator of the quality of the overall population and how much thought we put into our use of land creditors these days. It's certainly a far cry from some of those challenged locations that caused some of the pain back in 2008. Back in May at the Capital Markets Day, we reported that the pension scheme was fully funded on a technical provisions basis. That position was largely maintained through the quarterly funding updates in June and September. The combined impact of weakness in the equity markets in Q4 and the GMP equalization reduced the funding to 93.9% at the end of December.
As that funding level dropped below 96%, this has meant a resumption in contributions to the scheme from January. We anticipate a total payment in 2019 of GBP 47.1 million, which is an increase of GBP 13 million year-on-year. We were very pleased with the cash generation in 2018. We achieved 93% cash conversion, which is close to the top of our target range. This slide shows the actual cash generated by U.K. operations both before and after land spend, over the last five years as volumes and profitability have increased. Over that period, the U.K. business generated over GBP 6 billion in cash before land spend. I'd like you to note that GBP 1.4 billion generated in 2014. Pardon me.
As it approximates to the starting point for the analysis on the next slide, with the GBP 44 million difference between the two figures being the net impact of pension contributions, depreciation charges, and working capital movements. The two tables on this slide are a simple but hopefully effective illustration of why this business continues to generate significant amounts of cash in a downturn. The first table shows the cash flows of an average U.K. unit. The numbers in the column on the left tie into those presented in the U.K. margin driver slide. The columns to the right apply various price reductions and demonstrate how each unit sold continues to generate substantial cash contribution, even after, say, a 20% reduction in price.
The second table takes the results of the first table, applies them to a variety of volume sensitivities, then deducts net operating expenses at a fixed 2018 value of GBP 196 million to illustrate the cash flow generated by U.K. operations before land spend for each of those scenarios. The unwind of land creditors would, of course, need to be considered, too. As you can see from a slide in the appendices, GBP 353 million of the year-end U.K. creditor falls due in 2019, just under GBP 200 million in 2020, and the balance of about GBP 200 million over the following four years. Perhaps slightly more helpfully, in the unlikely event that we were to stop entering into land commitments as of today, then I would expect the total 2019 U.K. land spend to be approximately GBP 400 million compared to the GBP 612 million, pardon me, in 2018.
Couple of other things to bear in mind with these numbers. You would expect in a downturn to generate build cost and overhead savings. None of that is factored in. You would also get a cash inflow from the net reduction in your WIP and creditor balances. Again, none of that is included. You will see that the 10% volume and 10% price scenario has been highlighted in red, and Pete will touch on that later in the presentation. Although this is a simplistic model, it is very easy to see, even with some pretty cautious assumptions, why we have confidence that the ordinary dividend will continue to be paid in the event of a normal downturn.
Based on our experience of the last downturn and applying that experience to more sophisticated modeling, I'm very happy to reiterate that even in circumstances involving a 20% drop in price and a 30% drop in volume, we would expect to continue to pay the ordinary dividend, and Pete will cover the special dividend later in the presentation. Lastly, just to confirm for anyone who may have missed it, we have declared a final dividend for 2018 of GBP 125 million or GBP 0.038 per share, which is in line with our ordinary dividend policy of paying approximately 7.5% of net assets and not less than GBP 250 million per annum. This, combined with the GBP 350 million special dividend declared in May of last year for payment in July this year, means that we will return GBP 600 million to shareholders in 2019, subject, of course, to shareholder approval at the AGM.
All of which leaves me with a bit of a conundrum on which particular superlative to use to describe the declared dividend for 2019, which, if you hadn't already noticed, has a yield of 11% based on today's share price. I don't intend to read the words out on the slide, as I think you're all very much aware of the five-year targets that we set out in May of last year, and Pete's already touched on those. These results already achieve the margin and cash conversion targets and make very good progress towards that 35% target for return on net operating assets. The targeted reduction in the relative size of the land bank will be achieved through a number of different factors, not least an increase in volume.
As we've set out previously, there is potential for the volume to increase in 2020 and 2021, subject to how the market plays out over the course of this year. To sum up, a record performance in 2018 has left us with a very strong balance sheet. We're making good progress with the cost and efficiency program, and we're confident that the business will continue to generate strong cash returns in the future. I'll pass to Jennie.
Thank you, Chris. Good morning, everybody. I'm going to jump right in the interests of time. I think probably the important thing to say about our new KPIs is individually, they're each very important, but together, they will deliver our strategy. There are two measures on the slide that I do want to draw specific attention to and give a bit more of an explanation. The KPI that we're looking very closely at at the moment around build quality is the construction quality review measure. This is a key metric for us in measuring build quality, and it's assessed independently by the NHBC and helps us to monitor and target improvements in the build process.
The CQR inspections take place across all build stages present on the site that's been inspected, the visit is followed by a site-level report, which identifies areas of good practice and, of course, areas that may be in need of attention. 2017 saw our first CQR period, and we were placed 12th nationally out of 27 with a score of 3.74. The reports that we receive are data rich and are reviewed regularly by our site management teams. They're assessed for trends and continuous improvement opportunities. We've generated a cross-functional improvement process and undertaken modifications of technical details, targeted training, product, and sometimes supplier changes as a result. These causation and solution practices, I'm pleased to say, have been very quickly embedded, and in 2018, we saw a meaningful improvement in our score to 3.93 and moving to fifth nationally out of 27.
Another KPI just to draw your attention to is the direct trades. Through 2018, we progressed a number of pilots with direct trades. We've now developed a direct trade plan across all of our businesses. This is a more refined and defined definition than we used previously. It specifically refers to five key trades, bricklayers, scaffolders, joiners, carpenters, and painters. Following the lessons learned from the pilots, we're now using more of a hybrid model, which is a mix between mature skilled trades and apprenticeships. I know that Pete's going to refer to that in more detail later. Moving on to the more familiar. The positive benefits of our investments in customer services can be seen across a number of measures, but also by the fact that over 90% of our customers would recommend Taylor Wimpey to a friend.
A five-star performance in 2018. However, we are continuing to challenge ourselves and our teams by increasing our focus on wider measures of satisfaction. In this, we're looking towards the nine-month scores. Arguably, a much more complex influence and dynamic because it's assessed over a longer period of time. It takes a number of approaches in combination and sustained over time to meaningfully move this score. You can see that there are a number of contributory factors to the nine-month score. For quality, the comments typically refer to the quality of the finish and how long it takes us to resolve issues. The service after, and that time taken to resolve matters, are an important focus for us to address.
We're tackling this in a wide range of ways, including the build right first time, so that we have a process of reducing issues through the build stage, through our consistent quality approach, which we've spoken about before, the HQI process, which is now very well embedded, smoothing the overall build programs, and thereby delivering quality homes on time. The other questions, such as development, are also achieving quite a bit of work. These are closely overlaid with the customers' comments on quality. In that, we're continuing our work on design, evolving our new house type range, and enhancing our placemaking skills and activities. Looking at build quality, getting it right first time then continues to be a key priority in our customer-centric approach. It sounds really basic, but with so many variables, it has proved quite difficult for our sector to achieve.
Our approach, I think, is fundamentally different because we are taking a holistic approach across the whole range of the process. Right first time is good for our customers, but it makes good business sense too. There are material savings to be had by getting it right. In addition to the quality improvements that we expect from those processes like CQR and HQI, right first time should also enable us to be more efficient, to deliver more predictable build times, generate cost efficiencies in the long term, cut waste, and become more sustainable. Other initiatives that will help us deliver are, as Chris identified, increased use of the standard house type range and further standardization of the actual build process. The CQA process also continues to evolve, setting very clear and identifiable standards around key finishes.
This is now very well embedded in our customer service and build teams and is also now being adopted by our procurement teams as a quality tool for key suppliers. The next step will be for us to move towards a customer-facing version of the CQA, which will increase transparency for our customers and ensure that they know the quality of the build that they can expect from Taylor Wimpey. I think, as you can see, there's a lot going on, but it's not just one thing, it's a combination of efforts. I think we're doing well, and we feel strongly that getting this right will make us a more resilient and valuable business. Moving on to land, and I'll move through these quite quickly. We have had a broadly replacement approach to our short-term land bank, which stands at almost 76,000 plots at 5.1 years.
We've maintained a measured and disciplined approach to opportunities with the focus remaining on quality, adding over 8,800 plots at a contribution margin of around 27%. The planning environment has remained broadly positive, albeit the greater proportion of opportunities, particularly those at scale, are now delivered by or in alignment with the development plan process. The visibility, therefore, on the timing of consent drawdowns does remain challenging. We had a good performance in terms of the conversion to short term from our strategic land bank of over 7,500 plots and added over 10,000 plots to the strategic land bank. 58% of our completions this year were from the strategic land bank. A very familiar graph for you. I think it demonstrates that the businesses are continuing to select opportunities with strong margins and very pleasing return on capital.
The uncertainty over Brexit, particularly towards the end of the year, did enable us to push harder in certain parts of the market. I would say that any further improvements on these margins are likely to be opportunistic and unlikely to be sustainable for long periods. I do, however, think that there are improvements still to come in return on capital. Just a whistle stop around the land market. I think the market continues to operate well, and we do continue to see good opportunities presented. It's notable that larger house builders opening new regions and regional growers can be seen to be compressing margins for smaller sites in the short-term market, whilst larger sites continue to attract fewer bidders.
Whilst requiring detailed assessment and more resource intensive bidding in this part of the market is more disciplined, TW, given our credibility and our technical capabilities, perform very well in that part of the market. Looking in in London, and as an additional overlay to the comments on the slide, it is particularly notable that since the elections last year, that the political and planning landscape has changed with much greater weight placed on community consultation and local views by politicians. This has added further uncertainty on top of the policy issues that I mentioned here. Moving on to the strategic land, I think we anticipate that the strategic land pipeline of new sites will remain broadly stable for 2019. Our teams continue to focus on opportunities identified through structured land searches, and one-to-one opportunities remains a very significant part of our strategic land business.
It's worth noting that there is noticeable pressure on minimum prices within options at present. A symptom, I think, of the unease of landowners and advisors on possible planning policy tightening around viability at local plan stage. Given the focus on Whitehall and Brexit, it's easy to forget that actually the majority of our planning decisions, both on short-term and strategically, are taken at the local level. There therefore has been a significant amount of business as usual out in our businesses. Worth noting, though, the devolution deals which span multiple local authorities are gaining momentum and will become more of a factor in the way that we plan our land investment as we go forward. We're starting to see greater autonomy expressed by those combined authorities, though unfortunately not always positive.
The Greater Manchester Spatial Framework, for example, is worth bearing in mind, where housing numbers have actually fallen and a significant number of green belt releases have been removed from the most recent consultation. Clearly Help to Buy, and the announcement last year is of interest. This table sets out the percentage of first-time buyers by government region based on our 2018 private completions, and then identifies the percentage of those transactions which would fall within the post-April 2021 regional caps. As we might expect, given the relatively low regional caps, the greatest impacts from unwinding Help to Buy are elected to be in the Northeast and in the Midlands. As an exercise, we also then rolled forward our forecast completions for the period 2021, 2023 based on the current land bank and mix.
During this period, without any other mitigation action at all, approximately 58% of our private completions would remain within the Help to Buy regional caps. As a result, we can feel reasonably comfortable of the prospects of the unwind. We will, of course, assess where pressure points might occur. We're reviewing acquisitions to ensure that there's flexibility, and where appropriate, we'll look at changing build routes and remixing and replans. The security from our land and planning perspective for outlets for 2019 is very strong. We continue to use our project management process, PMIP, to track and manage outlet openings, and this continues to increase the accuracy of our forecasting, and we continue to improve this. The depth of the land controlled is a reflection of our strategic strength across both operational and strategic land businesses.
I'm really very comfortable with the overall profile of the sites, and the sites to acquire in forward years with an acceptable degree of challenge for our land and technical teams. This is, of course, regularly reviewed. A balance is needed. The controlled but not yet owned and yet to be acquired represents both risk and opportunity. This element will give us space to maneuver and navigate a changing market, should it be called for as we go through 2019. The level of investment we will calibrate depending on how we find those conditions. The greater the land spend, the greater the expectation of growth from 2020 onwards. The lower the land spend, though more cash generative, will obviously limit that expectation. A very careful balance is needed, and it's something that the management team is focused on.
Finally, just looking at our profile of strategic land conversions. We've been very successful in recent years with an average of 9,000 plots per annum. The second half of the graph shows what the next few years of potential strategic pipeline looks like. I feel the need to reinforce that this is planning with a capital P, so there are always casualties. We overlay a factoring, which will be our best guess of allowances for slippage, changes in planning policy and the like. Some sites drift by a year, some might drift by whole development plan review periods. The art of factoring becomes more challenging the later the years. Future pull-through, I think, will continue to be lumpy, I'm only allowed to say that once, based on changing provenance.
This will, given the number of very large sites maturing in our land bank, though underpinning our future growth, distort conversion levels and the short-term land bank from year to year. We continue to invest in strategic land, and this investment, together with the continuing progress in development plan adoption and the government's continuing pressure for 300,000 homes, means that my expectation is that our strategic land conversion will continue to grow. Really to close, based on our strategy approach to increased efficiency, we have increased our sales rate on larger sites, and we have met this with build output on a greater number of factories. 307 factories on 273 outlets.
The analysis shows that the sales and build rates per site has increased the larger the site size, where we can create a sense of place, offer an enhanced customer experience, maintain stable site resources, where supply chain efficiencies can be maximized, and where there's less sales competition. In this year's analysis, we can see the beginnings of the differential. Consistent large site setup, optimal market-facing mixes, predictable build programs, visibility of future work generating consistency of build and supply chain. Means that wider market conditions permitting, this could be pushed further as our strategy continues to develop and mature. This increase in efficiency will drive sustainable growth from the existing land bank and is an important component of our strategy.
We can demonstrate the operational capability to drive greater volumes on larger sites at a consistent quality, our aim is to do this more consistently across the land bank. Thank you.
Thanks, Jennie. Jennie has given you a run through with quite a lot of depth in some areas of the processes that we've been working on, and picked out a couple of areas, particularly around build quality and around land, that perhaps give you more detailed information at this stage than we normally would. I'm going to stand back and try and go back to an overview and tie that together. First of all, about how we see the outlook, and my first two slides look at the outlook, first of all, short, medium term, and then longer term. Then restate and update our strategy, but then come back to how we see the investment case and then finish on how we see guidance for 2019, 2020, and 2021. I'll try and keep it relatively brief and high level. As I say, first two slides on the outlook.
I've split it into two, because as I sort of touched on right at the beginning, I do think you've got to split a series of short-term risks on how we see the market longer term. The two to some degree merge and could impact on each other. Actually, we could have very different environment in 18 months' time, and we've got to look at both. Actually, particularly when we come to land spend, what's very difficult at the moment is getting that balance right. If we didn't have the short-term risk, we'd have a slightly different take on the right thing to do, and trying to make sure we keep our options open, particularly for growth in 2020 and 2021, whilst at the same time not betting the farm, because in recognizing the short-term uncertainties is a tough balance.
Splitting them in two hopefully gives you a sense of that balance, because otherwise it all kind of merges together. Our short-term take is clearly the near-term uncertainty remains. I think we remain of the view that the risk to this business from Brexit is largely around the general economy and around general confidence. We take a lot of comfort from the fact that we've seen two years of trading where that's been in the background and increasingly noisy over the last six months, and as you've seen, trading has remained strong. I think probably the single biggest factor which we wouldn't necessarily have expected two years ago, which I touched on earlier, is the strength of the lending supply, the resilience of banks in lending, and the very clear and quite strong signals about where lending is likely to remain in most scenarios.
We should take quite a lot of confidence from that, because without that, we'd have a very different view of those short-term risks. I think as Jennie has touched on, that also creates some opportunities, particularly on larger sites in the land market, both to pick up margins, which we see as risk insurance more than anything else. If the market remains stable, gives us upside two, three years out. I'm not going to go through all the points that I've listed at the bottom in terms of limiting our risk factors and opportunities on the bottom of the slide.
I would just pick on the limiting factor in a way at the moment is more related towards developer confidence, as in, are we really sure that we want to push ahead with our plans as aggressively as we might do in other scenarios, rather than actually what's happening out in the marketplace. Then if I move on to the longer-term outlook, I'm probably slightly more bullish than most at the moment. It's stating the blindingly obvious, but it's actually very important to our market. Underlying housing demand is likely to remain above supply in almost any scenario you can construct over the course of the next 10 years. Even if build steps up meaningfully, that fact will still remain in the background. I still see the risk against that, as I touched on before, is very much around interest rates.
I think that sort of gets pushed into the background too much in the conversation at the moment. The reality is the Brexit process has prolonged the length of time we have seen and expect to see low interest rates, and it actually kept the market subdued but stable for a longer period. Net at this point, it's been a positive to a very cyclical industry rather than a negative, because prices are incredibly stable. I didn't touch on prices in the short term, but I would say flat is the right word in every respect. Flat doesn't mean down, it just means totally flat across the board pretty much. I do think, and this is key to our strategy, that what we've seen over the last five or six years from a political point of view is likely to continue.
Although land is not an easy area for us, the overall supply of land with planning and with the potential for planning is likely to remain good compared to historic norms and is likely to remain ahead of us and the ability of the industry to process it. I'm using process to describe everything from taking it through the planning system at the beginning through to actually finishing a home on site at the end of the day. The industry's capacity is limited. It is still probably our biggest limiting factor, and that is not going to go away. We need to do more to make that work for us. That is not a bad market backdrop for the next 10 years.
I'm not saying there won't be a downturn in the next 10 years, but those underlying drivers that have kept that market stable and given us upside potential probably remain a little bit more than people expect. Against that environment, what are we trying to achieve? These are new words, but there is nothing new in the strategy. It's what we've been talking about for the last nine or 10 months, and it's very similar to what we've been talking about in a more general sense for three or four years. The words are slightly different. What we're really trying to do is deliver a great performance now. What we're really interested and focused on is how we build a better business for that longer term opportunity.
Even if there is a more meaningful downturn in two years or three years, still we believe those underlying conditions are there for the longer term, and it's worth us investing in them to remove some of the bottlenecks and to make the business stronger and better. We believe what we're saying on customers. It's nothing to do with the reporting over the weekend. It's nothing to do with hitting a five-star customer service measure. It's about an underlying belief that our industry needs to be more focused on its customers than it ever will be. We have a changing environment, both politically and with our customers, and that environment has been changing slowly in the background for 10 years. In the first five of those 10, we were slightly asleep to it. I think for the last four or five years, we've been focused on it.
We're still not getting it totally right. Jennie touched on some of the detail of what we're doing to try and make it better across the board, but that really is going back and retooling every aspect of the business. I think that has to be true across the sector. Some are doing it to some extent. Some are not doing it at all, but it is a major shift. We do believe it's the right thing to do. It will lower risk in the long term. I think you see more of that over the last two or three years after we started talking about it than necessarily you believed you would at the point when we did. I think it is important to understand where we are limited by resources. It is a huge motivator for our changing employee base.
By that I mean our employee base is getting significantly and progressively younger. Their mindset is different, and their attitude is different, and they get motivated by different things. We see a huge step up in the internal morale and commitment to the business and retention from that shift in focus. We do believe, and this is new words, and it was there in our thought process back in April when we announced our strategy, but perhaps we were a little shy about it. We do believe that over that medium to long term, we can create a growth potential that's greater than anybody else in the sector. That's the payoff at the end of the day for investors. Not in 2019 and probably not even in big numbers in 2020 and 2021. We do believe we can create something different.
Just standing back slightly on the customer piece of that, what does that mean in the short term? I don't want to spend a long time on this, but I just want to give you a quick sense what we're focused on right now. This is not the whole set of things we're doing. It's what we see happening in 2019. Jennie touched on making sure we get the basics right. We've seen a huge improvement in the final finished quality of the houses that we hand over. When we talk about right first time, we mean all the way through the build process, so we don't have to do so much internal checking before we hand the house over. It's not so much about what the customer sees, it's about the process behind that.
Our customers are telling us, "Until you get that right consistently every time, then we're not actually that interested in some of the fancier things you might do. That's what really matters to us." What they're also saying is, "You can help us, and we really care about the community that we move into. We care about who our neighbors are. We care about actually having forums that are set up, that actually you help us get a community feel to the environment that we move into from day one." It surprised us how strong that is and how much actually they believe we can make a difference. We think there's a huge sales benefit there if we get it right. We've got a lot of work going on around how we make that work.
I think from a value-added point of view, we're retooling every element of our placemaking, really challenging our businesses, particularly on these large sites, to think from the beginning, not just about what the development should look like, but when it should look like that. How we put in public open space and play areas and those things. How we get the development up and running quickly from a place perspective. Looking at a new house type range with more added value. Our house type range is good and solid and high quality, but if I'm honest, a tiny bit dull. It's moving it on and being a bit more creative. It's about enhancing the digital interaction with customers from options online through a whole series of other detailed processes. Last of all is looking at broadening the routes to market.
I'm not going to spend a lot of time on it today, but I will touch on that sort of deal that we did in the first eight weeks. That was not a deal where we had stock or a concern about sales on a site that we had developed and had remaining sales to make. That was actually about a new land acquisition where we effectively have taken risk out of the system to enable us to buy a larger piece of land and has a significantly positive impact on return on capital. If you look to the metrics, you will be very comfortable with it against our hurdle rates. What is interesting, and I don't expect us to do a large number of deals that look exactly like that.
If you look at what we did at the end of last year and what we've done in the first few weeks of this year, we have a much greater open-mindedness to explore lots of different opportunities to find different routes to market for our product. A lot of that, in our minds, is preparing the ground for a post Help to Buy world where I think we're going to have to work harder to make sure that we can get our products into our customers' hand and help them in lots of different ways with the financing. Going back to what we're trying to achieve, sort of beyond the customer. Two big areas, looking to improve in areas we think there is real sustainable advantage.
I think, it's not lip service, being the employer of choice in our industry is where we already are. Again, we feel there is huge benefit to that in a resource-constrained world. There's a lot going on to improve our offer to our employees and improve the culture of the business. Secondly, removing some of those historic bottlenecks. Jennie touched on investing in apprenticeships, in direct labor, and investing in new management trainees. We've increased all of those, as I touched on earlier, by about 50% year-on-year, and we'll increase them further next year. Actually also looking at the processes, the induction, how we actually make that work for individuals. We see that as a game changer in the long term. Summarizing the investment proposition.
The first few of these hopefully is what you've expected from us over the last few years. A strong near-term performance, a significant dividend. I'm not going to repeat Chris's sort of statement of what the yield is. A strong balance sheet, decent efficiency, real ongoing cash generation potential. Open, honest, and full communication. We always tell you how we see it and what's behind that. You start to get into areas where we already think we're decent, but we can get better. Better relationships with customers, better relationships with communities, better relationships with policymakers. Better efficiency, both cost and site efficiency and production. Better consistency so that across the board, across the year, everything is happening the way it should rather than it being all there at the end. The next one is quite a big one. A bigger ability to respond quickly to market opportunities.
If I look back at the last sort of three or four years, the thing I would be self-critical of is we were not able, because of production limitations, to respond to the market opportunity that was there. We had the land, we had the sites, we had the capital. We just couldn't step up quickly enough. Actually, we should have been in a stronger place next time around, whether that be in 12 months time or in five years time, I want us to be able to step up more quickly. Ours is a cyclical business. We have to be able to adjust, but I want us to have that capacity. That, combined with the ability to manage large sites better, gives us that bottom-line growth potential in the long term.
We do think, you put all that together, the potential to create a very different house building brand from what the sector has historically seen. I'll come back, as Chris touched on, to the special dividend, the scope for increased future cash generation, not just more of the same. Touching on that special dividend, I think, conscious of time, I just want to give you a sense of how we see the special dividend. Our internal discussions, plans, forecasting, as in almost all scenarios, the special dividend sort of in 2020 and 2021 and beyond, at the 2019 levels, plus inflation, and I'm not going to sort of spend too much time arguing about exactly what inflation is. That GBP 978 million kind of cash generation. We're not placing a great deal of store in a 10% volume, 10% price scenario.
It's an example, but that was on Chris's slide. That GBP 978 million of cash generation in that environment covers our short-term land requirement and that special dividend, and means we then retain the balance sheet strength to be able to invest in land. Even in downside scenarios, we still think a material special dividend is likely to be paid. I'm not going to sort of get into any more analysis on exactly how much and exactly what scenario it would in, but just that sense that we see it as a core part of what we expect to do over the next three or four years. The last point, and I don't want you to read too much into this, but we would never rule out share buybacks.
Clearly, when the share price was GBP 1.32 just before Christmas, as well as buying some shares myself, as did the chairman, we were seriously asking that question. I think with the movement in the share price, it doesn't feel as pressing. I would say that we would not expect to take a dividend that has already been announced, or strongly indicated, and not then pay it as a cash dividend. It would have to be either incremental or looking further ahead than that as part of a longer-term strategy, just to give you a sense of our thoughts on it. Overall, and particularly weighted towards guidance, we can't pretend the short-term macro outlook isn't uncertain, but trading is very good. We do think all sensible risk mitigation is in place. We're still relatively cautious. We still see this as a cyclical industry.
We will not bet the farm for short-term gain. We did pull back our land purchases at the end of the year. I repeat the comments I made in the trading update that our 2020 and 2021 growth will, to some degree, be dependent on our confidence in 2019 and whether we see through land and work-in-progress investments in the latter half of the year. The potential for growth in 2020 and 2021 remains meaningful and remains intact today. It hasn't changed over the course of the last two to three months. Future margins show a balance of pressure and upsides. What I mean by that, if you look at 2019, flat pricing, a little bit of material inflation, sort of some downside pressure.
Look into 2020 at the moment, even in the same environment, bit of upside from the improvements in land acquisition margins, slightly different sort of mix of business. More of a positive contribution from Central London, which is sort of pretty neutral in 2019. That affects the year-on-year piece. Probably a bit of upside into 2020 and 2021. Broadly, sort of in line with 2018. Again, I go back to our aim is to deliver a strong performance now and create significant future value potential. Questions, please, and feel free to address questions directly to Jennie and Chris. I don't need to take them all sort of through here. We should start at the front and then work backwards.
Morning. It's Gavin Jago at Peel Hunt. A few topics if I could, please. First one's just around the sales rate, what you've seen in the first couple of months. I think you've outperformed the peer group that we've heard from. What's differentiating you from them? Do you think it's mainly the large sites that you're operating from? You haven't got the granularity from all your peers, but just trying to dig into kind of why you think that sales rate is particularly strong. Second one's around Help to Buy. I mean, you've touched on it a little bit, Pete, but just how are you planning for life without it? Obviously got first time buyers, about a third of your completions. Really your thoughts on whether you think that industry volumes can continue to grow in life without Help to Buy.
Finally, just on the land spend, probably one for you, Chris. You gave an indication of that, but with all the larger sites coming through, is there going to be an increased WIP spend this year, and just a bit of guidance on year-end cash would be useful, please.
If I take the first three, then you take the last one, Chris. On sales rate, why? I'm afraid the answer is sort of boring because it's everything we've said before. I still think the most important factor is the land quality metrics that Chris put up before. Our mix of sites is very good. We have very few sites that aren't in places where people want to own a home. Particularly in less certain times, less confident times, that becomes critically important. We've argued for a long time that sort of when times are good, actually the more secondary markets still sell very well. When times get a little bit uncertain, then those customers lose their confidence more quickly or have less ability to buy. I think that's the biggest factor. Everything else comes down to right product, right place, right people.
Slightly less competition because we haven't got the level of site level competition that we have had in the past and that perhaps some of our competitors have. We're less likely to parcel up those larger sites into a series of smaller ones. We don't have two brands sort of internally competing with each other. To be honest, if you put all that together and we weren't generating better sales rates than the average, we'd be pretty disappointed. There is a substance behind the strategy. I think you can see it at the ground level very much. Probably, all of the things remaining equal, the phasing of Help to Buy through to 2023 is unlikely to have a material impact on industry volumes. It might do at a regional level.
Jennie pointed out, particularly in the Northeast and to a certain extent in the West Midlands. I suspect if others put up the same chart for you in terms of price gaps, you would see bigger gaps there in those markets because, it's about where your relative regional positioning and strength is and where you've tended to drive profits and volumes from over the last few years. Still, overall, there'll be winners and losers in that, and there's enough flex, there's been enough time for people to remix and balance that. I don't think that has a major impact on industry volumes. I think it is, if nothing else changes, removing Help to Buy completely probably does have an impact on industry volumes. I think if confidence is good at the time and remains good, I don't think it's huge.
I think it's very difficult for the industry to continue to grow in the short term through that transition. We can do all the mitigation that we've touched on around how we get product to our customers and how we support those marginal customers through that process and how we think creatively about those routes. I don't think it's completely compensatory. I think there is an impact there. There has to be. Chris, do you want to pick up the last one?
Yes, I think there were three, WIP, land, and cash. On WIP, I would expect, as Pete talked about, looking forward into 2020, obviously it depends on what happens this year, but in order to drive that growth, you would be investing a little bit more in WIP. No more than a couple of % is our expectation. Land, at this stage we don't know what the land spend is going to be for this year. It's massively influenced by what happens over the coming months. That obviously then is the biggest influencer on the cash balance at the end of the year. If we spent exactly the same amount on land this year that we spent last year, then clearly the cash balance would increase.
Do you want to just hand it to Will next to you, and then we'll move back to Glynis after that?
Is on. Yeah. Thanks, sir. Will Jones at Redburn. Three as well if I could, please. The first was just exploring the land-buying metrics you gave in the year versus the P&L. Land costs through the P&L were roughly 16% of sales. Your contribution margin, I think, was about 26%, round numbers. You bought land in the year at 19% of sales for an estimated contribution of 28%. It's probably inverse relationship to what we might usually expect. Firstly, I guess within that, are you absolutely sure of the 28% that, I guess, the teams feed up to you? Secondly, why that differential, I suppose, please? The second one, Jennie, I was intrigued by your comment around the possibility to improve return on capital land buying, even if not the margin. Obviously asset turn being the difference.
If that's the case, what's helping that? Is it the, I guess, the savings drive the groups putting forward in terms of, I guess, the timing point that was made around number of days saved? The last one was really just double-checking. In terms of the sales rate, are you absolutely happy that you are in no way foregoing anything around the price side of things to deliver the sales rate we're seeing? Either way, is there any argument from here, I suppose, just when you look at the order book and where you've got yourself, too, that you might just pivot slightly towards margin for the rest of the year, considering you're in a strong volume position? Thanks.
If I pick up the sales rate and margin question. You want to take the land ones, Jennie?
Yeah. I think around return on capital, there are a number of approaches. The actual structure of the deals, and we're seeing that particularly with larger sites and some softening or a lack of other active participants in that part of the market, that we can drive much better structures in the overall deals. Probably the bit that I'd also focus on is all those improvements that we're driving through efficiency, taking out as much of the friction in the build processes as possible, driving up our sales rates. All of those are additive to improve return on capital. A little bit in the land deal structure, but quite a bit I would expect that we can drive just by the way that we're delivering our sites.
Yeah. Yes, we're confident in the 28, Will. On sales rate, are we giving up price? There's always a trade, isn't there? There's always a trade between price and volume in any market. No, on a relative basis, either relative to peers or relative to history or relative to what we expected as we set our plans for this year, no, we're not. We always have an allowance for a certain level of incentives in how we plan out the sales tactics at an individual site level. That hasn't increased or changed fundamentally. You always have some sites at a local level where you win a bit and you lose a bit, and that's in the overall flat measure. It actually is very, very stable across the board. We're not chasing those numbers in a way that is taking an adjustment on price.
I'd be very surprised if our price performance is different in any meaningful way. I think, there isn't a huge upside opportunity on price, even if we were to take those volumes. Because people are cautious about what is the right price. So, it isn't like even if we cut sales rates by half, prices might move by half to 1%. They wouldn't move by four or five, if you see what I mean. The level of elasticity is relatively low. Will we pivot more towards price over the course of the next few months? I think the answer is yes and no. From a medium-term point of view, taking the year as a whole, no, I don't think we do.
It is normal for us to want to spend the first quarter getting the order book in a strong place, giving ourselves choices for the year, not to be the last one trying to sell the last house for completion in October or November. That's always true. Yes, later in the year, the balance will probably be a little bit more weighted for price if all those conditions are the same, but only in a normal sense. If we go through the current political shenanigans, I'm quite proud of getting the word shenanigans into this, if we get. It was the most benign word I could think of, to be honest.
If we go through those over the next few months and we have a deal and we have an exit process, is it possible there's more price upside at the back end of the year? Yes. I mean back to a normal, underlying inflation, 2% pace. Yes, I do think that's perfectly possible, but probably too late to really impact on this year in terms of completions will be my feel. Our base case is flat pricing through this year in terms of a view of margins for this year. Cool. Do you want to pass back to Glynis, and then we'll come back to the front again?
Yes, morning. Glynis Johnson, Jefferies. Again, three if I may. Three is the magic number. The first one is in terms of the construction quality review figures that you gave us. Can you give us a little bit of color around them? What is the highest number that actually is achieved? Is government focusing on it? Is there a risk that they turn around and say, "We need you to get six"? If so, what does that mean in terms of costs? Second one, in terms of the strategic land slide, so I think it was slide 35. It seems to suggest a very big pickup in the later years. I'm just wondering, are there one or two very large sites that skew that? Clearly the lumpiness, and I got the word lumpiness in, that could have a big influence.
Lastly, in terms of, well, it's a question that you always don't like, but outlets, and I'm going to reference it to factories. On your super large sites, that's slide 36, the number of outlets versus the number of factories. I was quite surprised. It seems to indicate that a number of your very large sites only have one factory. How should we see that progress? Should we be assuming that actually your very large sites should, on a two, three-year basis, have two factories, three factories? If you could give us a bit on that.
Yep. I'd just like to make a comment first on the CQR then sort of let you come back and cover that as well, Jennie. Probably on the strategic land as well. I'd sort of make a comment then sort of let Jennie pick it up. On the CQR from a government point of view, no. I'm not sure that anybody in government would. Actually, probably except for the ones we've mentioned it to. I'm not sure anybody else in government would recognize and know what it's about. Similarly, that shift from the eight-week survey to the nine-month survey, I don't think Jennie did touch on, but we're also looking at net promoter scores. To me, it's what I mean when I say open and honest communication.
We are telling you the things we think are important and that, in time, other people will start to look at, if you see what I mean. It's not a reaction to government. It's actually for the first time on that CQR thought, we have a usable benchmark for our sites on build quality, and there's never really been that. We see that as a useful tool to improve, not because somebody else is looking at it and we're worried about the score. I mean, I'll let Jennie pick up on what we think is a reasonable score overall. It's because we think it's a good and useful way of understanding how the business is performing, and it's important to us, and we think it tells a story about what the business is doing. That's generally true across the board on some of those metrics.
They're not reactive reasons we're putting them up. They are, we think this is going to be where the debate is over the course of the next five years. We'd rather be ahead of it. Just on strategic land, I'll give Jennie the chance to answer sort of on both of these. Just on strategic land. There is, you'll have spotted it, Glynis, there is a sort of a natural tension between shortening our land bank and more strategic sites. I see the shortening the land bank as about balance sheet efficiency from a capital point of view.
I am reasonably indifferent if we happen to have a number of very large strategic sites that we haven't paid for and aren't fully committed to from a price point of view, and they sit in the land bank and throw that metric out but don't impact on capital and we're running them efficiently. That's not an issue sort of from my perspective at all. The signal about shortening the land bank is we can make the sites we've got committed with capital out much more efficient. If that means tweaking the metrics to get that balance right because the proportion of strategic, that's a good quality problem to have, if you see what I mean.
I recognize it creates a tension in the metrics, but underlying, if we can do both and shorten our time on sites where we've got capital locked up and get better growth out of them and have lots of large sites coming through the strategic land bank, then I'll live with that problem. Sorry, Jennie.
Yeah. CQR, it's not a measure that I think government have any visibility. The reason that it's attractive to us is because it takes a whole site view. It's not the plot inspection. When the NHBC come to look at a plot inspection of a home, they're looking specifically at that plot. It's a much truer, honest view of the health of build on a site. I mentioned just how data rich the feedback that we get as part of the CQR is, and we've been using that to its absolute maximum to identifying where we think we have issues, where we can drive improvements. It also gives us a benchmark where we are nationally. I said that we were fifth. It's quite secret squirrel. We're not really fully aware of who's ahead of us, but I think that we are the largest volume builder.
We're sitting really well there. We'll be able to continue to drive that data into improvements, and it's exceptionally beneficial to our site teams and our site management processes to achieve that. On strategic land, it's always really difficult for me to put a slide like this up, but you've hit on the thing that makes me most anxious, which is things move to the right. Yes, there are some large sites in there. If a 3,000 unit site that we have identified for a local plan allocation in 2021 slips because the local plan has been stalled, then those 3,000 units are going to move to the right. These are our best estimates, and it's probably best to split them into sort of three sections. There are sites which we of course draw down.
There are sites that either because the five-year housing land supply that we think might be available would slip, or that sort of development plan overall, or the site doesn't get allocated. There are a number of things that happen there. Large site landing is something that we're really focused on. A really large site doesn't come out of nowhere, so we have preparation time. There's a lot of detailed technical work that our teams put in the transition and transfer of sites into the business. It's something that we can manage either in the way that we manage land spend, the way that we break up the phases and how we manage it, and give sort of custody into the short-term land bank. I'm pretty comfortable about that.
Remember, these are good opportunities that we've been working on through our placemaking, setting them up so that they're absolutely optimal for the time they land in our short-term land position. Then I think just on outlets, we're in transition. The point that I would make here is we're starting to see the differential, Glynis, but we're not there. It's a journey, and we've got a long way to go. This is also a record of the sites at their opening capacity. Some of those will be actually getting potentially to close, so it would be reasonable for there only to be one factory. It's an entirely dynamic, but I would expect to see that pulling away a little bit more as we go forward in the strategy.
Can we pull back to the front and then work our way back down this side, and then just conscious of your overall time, so we'll try and okay, that's fine. We'll try and keep our answers a bit tighter so we can get through everybody's questions.
Good morning. Jon Bell from Barclays. I think I've got two. Pete, in your analysis of housing booms and busts, one market that springs to mind there might be Spain. Just wondering what your current thinking is on that business. Then secondly, one consequence of your high sales rates, in recent times has been strain on your open site numbers. You've just reported another period of very strong sales rates. What practically can you do to replenish those sites?
Nothing new to say on Spain. I don't think it's about to bust, don't think it's in a boom either. It's kind of solid, but okay. There's certainly nothing strategic to talk about. In terms of open site numbers, it's there in Glynis' question as well. We're not uncomfortable talking about it. Actually, it's a totally reasonable question to answer. It is a pressure point within the industry and within the business. I think what I would say, and it goes back to Jennie's charts to a large degree, our site numbers are very resilient. They're larger sites. They're open for a long time, so we're less sensitive. We're quite happy with where our outlet openings sit over the course of the last few months.
I think I am right in saying that every single outlet we plan to open this year has opened when we plan to open it. I'm not sure I could ever have talked about any period in the past where that had quite happened. It is always something that's under pressure, but we feel that it's actually operating very well. I would still say it'd be quite nice to have another 20. I want 20 proper sites, not to stick 20 new outlets on sites we've already got or buy small sites that then create an issue within the business. We'd like a little bit more. We expect it to grow a little. We're committed to doing it in a way that will really add to the business rather than just window dress the number in the short term.
Okay. Thank you.
Can we move across that row, and then we'll come back to the front. It's all right. Let's go to the front since you're down here, and then we'll.
Thanks. Gregor Kuglitsch from UBS. A couple of questions. The first one, just to elaborate a bit maybe on the 2020, 2021 kind of risk profile to volumes. I guess if you could just sort of give us some sense what you need to do this year. For instance, if you do buy at 1 times replacement, do you think you can step up the volumes as you were indicating last May? Or would you need to buy more land just so we can get a feel what we need to track to get some confidence around that? The second question is, you kind of elaborated on the cycle and Help to Buy ending, and obviously you mentioned that volumes will probably come down. I guess my question is, what about margin?
Do you think that there will also be generally a step down in margin as a result? Perhaps incentive costs have to step up. I think you are preparing clearly for some of them, trialing some of those. You have some experience, perhaps, to share on what the margins are on sort of, if you want, free market sales, which do not benefit from a subsidy. Thank you.
Okay. 2020, 2021 volumes and land. First of all, Jennie's sort of chart about where our land supply sits for the next few years has meaningful, and I am not going to give you a number, but meaningful higher single-digit growth in it. Actually we have the potential, if we see through all of those acquisitions, which would result in a growth in the short term land market, to grow more materially. I think that means that the confidence in a stable volume for 2020 as a base we can work from is quite high. It is really hard for us to relate next year's volume to this year's land acquisition. Not that many sites we buy this year impact directly. It is the overall investment that it takes in us bringing forward phases and things.
It is really hard to give you a specific answer to that question that is not quite misleading. It is not just about land, it is about how we open up phases, how we invest in sort of the infrastructure during the course of this year, rather than just the sort of site numbers. On the margin impact, I think you just asked me for margin guidance for 2023. I think it is a given, do not you, that volumes and prices would see some degree of pressure in the industry, if everything else was equal, at the point when suddenly there is no Help to Buy. I think it would be naive for me to give you any other signal than that there will be some pressure.
I think we can do things to mitigate that. I don't think it's by any means necessary that it is a catastrophic change, but it would put pressure on in that year from an adjustment point of view. I think we have to get nearer to the time and see the conditions before I can go much further than that. I think it would be naive to guide you that that will be anything other than a negative to the overall industry. I think the big question and the thing that we want to make sure in the communication, the process and how we prepare is that enough of a shift that it creates real uncertainty in the market? Because having to accept that you can sell less houses that year as you adjust and you use other measures is one thing.
It's if that creates a sense that the housing market as a whole has changed. That's the risk that the government and we should really want to avoid. We can adapt and we hope to do a better job of it than others in adapting to that change. Is that change big enough to affect the market overall and the second-hand market and confidence? That's the most important question, I think. I do think it's a given that at that point of change, there is some impact on volume and prices, but I couldn't give you a quantification sort of four years plus out. Happy to deal with the question every time till then.
Even in three years.
Can we just pass it back?
Hi. Chris Fremantle from Morgan Stanley. Talked about sector leading growth. I just wanted to pick up on the special dividend policy again. I know you were both at pains to emphasize the dividend yield. It doesn't seem to have done much for your share price since you announced it, which perhaps is a reflection of, I don't know, the ability to grow earnings when you're paying out that level of today's earnings. Can you just give some thoughts, just remind us why you think that's the right thing for the business, particularly if you aspire to create sector leading growth?
Yes. The main answer, or the most sort of significant answer to the question is, we are not constrained by capital. At the moment, we have a series of other constraints, and I think we've tended over the last years to see land as a constraint. As I touched on, I think that's changed a bit. Yeah, I think a lot of our strategy last year was about recognizing that land was a different balance to what we have been used to, and that we should adapt our strategy to that. Our biggest constraint, as I've touched on, and I think comes across the whole sector, is actually our ability to take sites through the planning system from a detailed processing point of view, to get those outlets open, to get construction on site, to be able to match sales.
Actually, in a cyclical world, at times sort of the ability to sell, it's not capital. We genuinely think the special dividend is excess capital that we could only invest if we were prepared to significantly reduce the margins and returns that we're actually able to secure sites at. There's always another site that you can buy that's more or less the same returns. If we took sort of an additional GBP 600 million in capital and came close to doubling our land spend, we're big enough it would affect the market. Actually, I think they're very consistent. You go back to, we think we can drive more growth from a combination of the larger sites and as long as we can get the sales rate and the product delivery right on those sites.
From our strategic land bank, which is relatively low in terms of its capital absorption as we bring it through the balance sheet, without using a massive amount of capital. That's why we think there's efficiency, capital efficiency on the way through that. I certainly don't think it's inconsistent. It's just a different part of the value proposition. We can deliver significant growth and still create a fair amount of cash. Just a follow-up thought to that. Very different when sort of our land acquisitions are sort of high teens proportion of selling prices to if I roll the clock back 10 or 12 years when at the point in the cycle we're at now, they were more like 25% to 30%. Makes a big difference to the cash dynamics and the balance between growth and investment.
Hi guys. Emily Bédouet from JPMorgan. I've got two questions, please. The first one is just coming back on the sales rate. I'm conscious that you sort of started to post decent year-on-year growth in the second half of last year, and you're still sort of running at that kind of rate. Everything you're talking about at the moment to kind of drive sales rate improvement, when we get to the second half of this year, if the market looks the same, are you going to be sort of telling us to remember the comp stuff, or would you be alarmed if we kind of had similar kind of growth in our models for the second half? Then the second question, just in Q4, there was that one sort of big one-off deal.
There's something that's sort of slightly different, but another sort of sizable deal in Q1 this year. Is this just a completely random sort of London buses kind of thing, or is there something in the market that means we should expect more of these to come? Thanks.
If you take the full year sales rate, assuming a base case of the market remaining through the rest of this year, more or less as it is, I think my view will be, I would expect the sales rate for the year to be higher than last year's, but not to be 10% higher. The first quarter was a decent first quarter, but the comp wasn't massively difficult. You're right, it picked up at the back end of the year. I wouldn't expect to have another step up beyond that pickup in the last quarter. It would sort of normalize a bit closer to last year as we go through 2019. As I say, gut feel today, and don't take it too strongly, it is probably a bit ahead overall. Nothing we do is random and like London buses, Emily Bédouet.
I'm offended by the very idea. It's both. One often it's opportunistic, but I think it fits the overall sense that we're talking about routes to market. We are open-minded about looking at different ways of selling our product. Sometimes that's about, sort of as the deal at the end of last year was, across a number of sites, where there's the opportunity to build the order book and get ahead. Sometimes it's like the deal we've done in the last few weeks, which is about much more front end, about securing the risk on a site and letting us buy a bigger part of that site without feeling like there's a material sales risk on a significant proportion of it. If it works, we work well with the partner, and the numbers work.
On that latter deal, there's a very small trade-off on margin, but it still fits in our overall strategic kind of projections of margin, and a very big gain on return on capital and a very big gain on risk. We'll do them. It fits the overall strategy, but they're also opportunistic. I can't tell you whether we'll see two or three during the course of the year to come or none. You're more likely to see them in 2019 than you would have been in 2017, sort of for reasons of both the market and the strategy. Can we just move back, please?
Morning. Andy Murphy, Bank of America Merrill Lynch. Just two questions. You talked about investments in the statements gone up from 3% to 5%. Is that the sort of the bulk sales that you're talking about, or is that some signal that maybe buy-to-let investors are coming back into the market? Secondly, just can you talk a little bit about what you're doing and what the future is you think for the Springboard rental option? Sounds quite interesting.
Okay. Do you want to take the first one, Chris, and I'll pick up the second one?
I think the investment's on the balance sheet.
Right. Okay. Is that right, Andy? You mean the balance sheet investment?
The sales to investors.
Sorry, the sales to investors. I thought you said investments rather than investors. I mean, to be honest, both are small numbers, and we don't see it as particularly changed. It tends to be very London-centric, and we don't see it as a structurally growing part of the business over the next few years. It's movements in a small number, if you see what I mean. In terms of Springboard, we have rented, I was going to say sold them, which would be completely misleading, rented the 15 properties in our pilot scheme. That's gone well. It's a learning exercise. Probably later in the year, we'll give you a kind of granular view on how we think that's working, whether we think those customers are likely to then convert into sales.
We see it as a pilot to learn from. We'll probably do two or three other things that sort of are similar but with different characteristics. Again, a bit like the sort of scale sale that I was talking about with Emily a moment ago. We'll look at more things like that are alternative to routes of market over the next couple of years, nothing new on that particular pilot at the moment. Thank you. We'll go across to Aynsley.
Thanks. Aynsley Lammin from Canaccord. Just two from me, please. Firstly, wondered if you could comment on sales rates and pricing in London. Just interested in your views there. Secondly, just on the kind of the outlook you put forward for the medium term, it's very positive, obviously. Just wondered, have you changed your view at all of what you think the optimum scale of a house builder like Taylor Wimpey should be on a medium term view? You get through the kind of near term political shenanigans you call them. With the help of PRS and other deals, any change in view there, which I guess ties a bit into the special dividend comments you made earlier? Thanks.
Yeah. Sales rate and pricing in London. Jennie, are you happy to just pick on those, just as a general guide?
I think Chris has that.
Yeah, I'll pick it up.
Yeah.
For the 8-week period last year, the sales rate in the wider London and Southeast division was 0.77 versus 0.78. Actually, very marginally up. I would say that probably pricing, as Pete said, is very, very flat. I think probably more pricing pressure in London and the Southeast than the other regions, but it's sort of marginal. Yeah, that's the position on that.
Yeah. Have we changed our view on the optimum scale of a house builder? Yes, if you compare it to what I would have thought 3 years ago. Not massively in the last 9 months. It's not so much about that medium term outlook, because, mind you, I don't think that's changed massively in the last 9 months either. It goes back to land, and it goes back to the whole debate around strategy back in April. If there is more land supply and we have to compete less hard for every site, and there is that underlying demand, but the underlying demand hasn't really changed in that period, but the land, in our view of it, has changed over that period, then the natural optimum scale is bigger. Those are the 2 big moving parts we're always going to deal with, demand and land supply.
Land supply has changed over the last few years, and that's changed our view about where the balance should sit. Again, I go back to, that doesn't get away from us being a cyclical market or a cyclical industry, so you've got to be prepared to flex within that. Where the upper end of that bit, with more land at good returns, the upper end should be greater. Government will be pretty worried if I answered any other way, but I actually do believe it as well, which is nice. Yeah. Charlie.
Chris. Morning, Chris Millington at Numis. I just wanted to pick up on the sensitivity analysis you did on slide 22. You made the point there's no build cost reduction factored in there. Just from your experience of the past, what do you feel the relationship is with volumes and prices and cost reductions? That's just the first one. Next one may be for Chris, but land credit is up in the period. Most companies we've seen so far are actually seeing a slightly reducing land creditor profile, whether it's just from the land activities they're doing. How do you see that progressing going forward in light of the larger sites? If it does grow, will that be compensated for by a bigger cash balance from the adjusted gearing point of view?
The relationship, you're absolutely right. I think it was touched on in the notes on that slide, and Chris touched on it. If prices fall, we expect cost to fall. It's always hard to quantify how much, and probably more importantly in terms of cash, because that's our one-year number. How quickly, because the cost impact tends to take 12 months before it really comes through the P&L and cash generation. It's not quite as responsive as some of the other elements of cash generation. It's also quite hard to call in a very different land environment where land is proportionately less. How land prices change, you can't really use old metrics. They're not really very relevant.
Land will clearly still have a material value, will it actually sustain a residual value in the way that we're used to if the market were to change significantly? On build cost, what's reasonable to assume? 10%? Yeah. Order of magnitude. It tends to be taking out the last two years or so of inflation, rather than being able to reduce build cost by 25% or 30%.
Is that 10 on under the 20 decline in prices? Sorry.
That's 10% of build costs.
Under quite a severe recession, I presume.
Yeah, I think it's less because the limiting factor is the people and the materials and the supply chain and what the other choices those individuals and those suppliers have. That's why it's quite hard to answer, because I'm not sure that that build cost number is much different between a 20% price fall and a 30% price fall. It is more related to volume. I think in a meaningful downturn, is a 10% reduction in build cost reasonable? Yes. I'm not sure in a much more meaningful downturn, it suddenly gets massively bigger, if you see what I mean. I think volumes contract because it's harder to get the overall economics to work. I think we've got two more at the back. Sorry, I skipped past Chris.
Very quickly, I tend to look at land credit as not just with one measure. The absolute size of them, the adjusted gearing that you mentioned, land credit as a percentage of gross land as well, which I think we're around about 27%. You're quite right. If you look at our land credit balance, around about a third of it is, I'll use the lumpy word, spread over five sites. Yes, as our strategy moves towards those larger sites, we get more opportunities. I think where it goes in the future and our level of comfort will be a factor of looking at all of those three sort of views, if that helps. That's fine. Thank you.
Cheers. If we move back that way, then I think we've got two more people with questions, then we will close.
Robert Easton from Goodbody. Just on the whole thing about getting it right first time. When you look at your sites that get it more right first time, what is the prize that's up there to get in terms of build cost on a square footage basis? I know it's difficult. Every site is difficult. Just give us a sense of what is that prize and when you look across your sites. Just on the same topic, if you do get it right first time, is there an ASP angle to this as well, and it's not just a cost-driven?
Yes. I think there is absolutely a selling price, but probably in more substance, a sales rate impact of getting it right. The reason that it's slightly more rate than price goes back to the answer to, I think it was Will's question earlier. There's not a massive amount of elasticity on price for many of our customers anyway. People are buying the home that they can afford, and they will buy a bigger home given the choice. There is absolutely a sales rate impact. Probably a slight price impact, because the two are always related. It's very hard to quantify the cost.
The answer I'm going to give, which I'm sure Jennie and Chris will feel uncomfortable with, is actually more about overall consistency and when we see a business and a site getting it right consistently across the board compared to a business or a site that don't. It's not just about getting the build right first time. The impact is comfortably 10% of overall costs. Between average and best, let alone between best and worst. Consistency across the business and across the industry is not good and never has been. It's a hell of a lot better for us than it was four years ago, and we're very comfortable it's better for us than it is for most of our peers.
I think it's really important to understand our industry has grown up as being a fairly cottage business with every single site having slightly different practices. As we gradually standardized and improved, you see those benefits and the gap between best and worst is huge, which means the potential is significant. The reason it's uncomfortable is it's hard to get at and it takes time. It's uncomfortable because we know the prize is significant, but we really struggle to work out when can we get it and how do we get it everywhere. Is that fair, do you think?
That's fair.
Thank you. Then very final.
Ami Galla from Citi. Just one from me. You've touched a lot on the build process and getting more consistency in your business. I'm wondering, when you think about your medium-term outlook, is there a forward plan of investments in, say, offsite construction going forward?
Sorry, is that a forward plan?
When you look at your medium-term growth in the business, do you have a plan for investing in a factory build construction?
We don't have a plan, but we are open-minded. We've touched on this sort of, before. Are there elements of the build process that we could simplify, standardize to take offsite? There are elements. We still struggle, and we are still open-minded to find something that it's the metrics we need in terms of cost efficiency, adaptability to a whole series of different planning and local requirements. We keep looking. I would never rule it out, and we have invested in new people on the R&D side to really look at it. Where we tend to find the gains is improving the detailed processes and the individual components rather than one big sort of ticket factory-based construction. We'd never rule out. Sooner or later it will come. It still feels like it's a way off. Thank you.
Thank you for sort of both lots of, but also some very good questions. Thanks very much. Look forward to seeing you at the half year.