Ladies and gentlemen, good morning, and thank you for coming to Centrica's 2017 preliminary results presentation. This morning we'll be reporting on our 2017 results and also providing an update on the implementation of our strategy. I should forewarn you that this presentation will probably last about an hour and 20 minutes, slightly longer than normal, given a number of things that we want to address. First, a word on safety in this building. There are no planned fire alarms today, and any building evacuation will be announced by tannoy. Emergency exits are marked at the front and rear of the auditorium, and Goldman Sachs staff will be available to direct you to the muster point, which is towards the rear of the building on the junction of Stonecutter and St Bride Street.
Let me now provide a summary of where we stand and what we'll be addressing today before Jeff takes us through the 2017 results in detail. Our performance in the second half of 2017 was weak, particularly in business energy supply. This weak performance, announced in November at the time of our trading update, combined with uncertainty around our future prospects in U.K. energy supply, significantly amplified by the Prime Minister's announcement in October that the U.K. government would pursue a market-wide price cap of the standard variable tariff and other default tariffs, taken together, resulted in a significant fall in our share price in 2017, particularly over the fourth quarter. Some of the drivers were clearly beyond our control, but I regret the outcome deeply and the impact it has had on our shareholders.
I am determined to restore shareholder value and remain firmly of the view that Centrica's strategy can, and indeed will, deliver attractive total shareholder returns. We have excellent people, assets, and market positions, and are one of the companies shaping change in our markets at least as much as we are being shaped by it. Our focus in the near term is on performance delivery, managing through the political and regulatory uncertainty in the U.K., and maintaining the strong balance sheet which we've worked so hard to achieve. Although we delivered our 2017 published targets in terms of operating cash flow, cost efficiency and headcount reduction, capital discipline and net debt, our financial results in the second half of the year were weak, with a material miss in North American Business.
This was both in terms of the performance of our power supply book and from a charge relating to historical revenue recognition in one of our billing systems going back to 2013. Performance in U.K. business energy supply was also poor. Given this performance and uncertainty over our future prospects, I do recognize there are important questions arising from 2017, which we're aiming to address today. Let me come to these in a moment. In 2015, we embarked on a return to the core of Centrica, energy supply and services, and there are clear and encouraging signs of progress. At the end of 2017, phase 1 of Centrica's repositioning has been completed. We have materially repositioned the asset businesses and delivered on our GBP 750 million per annum cost efficiency program three years early.
We ended 2017 with a strong balance sheet, with net debt of GBP 2.6 billion in the lower half of our end 2017 target band and our credit metrics at levels in line with our target strong investment-grade credit ratings. As we look forward to 2018 to 2020, although we have still got some portfolio repositioning to complete, the focus is on performance delivery and financial discipline. In performance terms, it's about growing gross margin through our customer relationships and driving the next phase of cost efficiency. Today, we announced an increased cost efficiency target of GBP 1.25 billion per annum by 2020 relative to 2015, representing additional efficiencies of GBP 500 million per annum.
Despite the uncertainties we face in the U.K. energy market, we're also targeting to deliver on average GBP 2.1 billion-GBP 2.3 billion per annum of adjusted operating cash flow over 2018 to 2020 and to keep net debt in a band of GBP 2.25 billion-GBP 3.25 billion, designed to accommodate a range of scenarios and commodity prices, and being consistent with maintaining our strong investment-grade credit ratings. In terms of the dividend, we have indicated since 2015 that it would be linked to operating cash flow, and we expect to maintain the current level of dividend over the period 2018 to 2020, subject to meeting these operating cash flow and net debt targets. I'll return to forward guidance later in the presentation. Let me now turn briefly to 2017 financial performance. Adjusted operating profit at GBP 1.25 billion was down 17%.
Earnings were down 22% at GBP 698 million, equivalent to GBP 0.126 per share. Adjusted operating cash flow was GBP 2.07 billion. EBITDA of GBP 2.14 billion was down 9%. Centrica Consumer delivered robust performance with adjusted operating profit down only 1% despite the impacts of warm weather, the U.K. government prepayment cap, competitive intensity, and investing for growth. Centrica Business saw adjusted operating profit down 67%, reflecting the very poor performance in the energy supply business units. I've already covered our efficiency program delivery and net debt at the end of the year. Finally, we announced a full-year dividend of GBP 0.12 per share, unchanged on a year earlier. I recognize that 2017 has given rise to a number of questions about Centrica's performance and uncertainty about our prospects in the face of stiff competition and political and regulatory intervention in the U.K.
After Jeff has presented the 2017 results, I'll provide a strategic update designed to address the following questions. We know these questions are on the minds of a number of you, and it should not surprise you that I've also been reflecting on them. I'll review our strategy and how the component parts combine to reinforce each other as we deliver for our customers. This will include looking back at phase 1 of repositioning Centrica and the priorities we see for the next phase. Delivery of cost efficiency has been a material part of underpinning our performance over the last three years, and we will outline the role it will continue to play in offsetting competitive pressures on gross margin. I will address the resilience of Centrica in the face of strong competitive pressures and the specific issue of a potential price cap in the U.K. energy market.
Although there remains uncertainty, I will explain why we believe the actions we are taking will improve the market and allow us to maintain a healthy and attractive business in U.K. Home energy supply. I'll cover the customer account losses in Consumer, our focus on value, not volume, and the relationship between customer numbers and gross margin. I'll also describe how we're doing in growing gross margins through customer-facing propositions other than energy supply. There are some very encouraging indications of growth potential. Jeff and I will share our findings on the response to the issues in North American Business we encountered in the fourth quarter of 2017, and why it remains an important part of the portfolio. We will cover our intentions and the outlook for both Exploration and Production and our nuclear shareholding.
We'll also address the capability we're building for the future needs of our customers and our attitude towards acquisitions and disposals before finishing on our financial framework, sources and uses of cash flow, and 2018 to 2020 guidance, including the outlook for the dividend. Taken together, I hope this will help substantially in answering many of the big questions on the minds of those who follow Centrica. Let me now hand over to Jeff to take you through the 2017 results.
Thank you, Iain, and good morning, everyone. Before getting into the financial results, let me first start with the backdrop of commodity prices and weather. First, commodity prices. Brent oil, NBP gas, and base load power prices were all significantly higher on average than in 2016 and remained in the band between the 70/50/50 scenario from the time of our 2015 strategy review and our 35/35/35 low case. With respect to weather, temperatures in the U.K. were warmer on average compared to 2016, resulting in lower energy consumption. In North America, temperatures were slightly colder. However, more significantly, there was limited gas price volatility, which reduced optimization opportunities for North American Business. Let me now move on to the financial headlines.
Revenue was up 3%, which included the full year of Neas Energy revenues following its acquisition in October 2016, partially offset by the impact of lower average customer holdings compared to the prior year. As you just heard from Iain, adjusted operating profit fell by 17% to GBP 1.25 billion, reflecting significantly reduced profit in Centrica Business. Adjusted earnings also fell to GBP 698 million, including the net effect of higher interest costs and a lower effective tax rate driven by business profit mix, U.S. tax rate changes, and tax provision releases. Adjusted basic earnings per share was GBP 0.126, and the full-year dividend per share is GBP 0.12. On cash flow, EBITDA reduced by 9% to GBP 2.14 billion, in line with the reduction in adjusted operating profit.
Adjusted operating cash flow fell 23% to just over GBP 2 billion, inclusive of the one-off working capital inflow in UK Business in 2016 of GBP 357 million. Adjusting for this, and for foreign exchange and commodity price movements, underlying adjusted operating cash flow fell 13% compared to 2016, and on a cumulative basis is now broadly flat since 2015. Group net investment, which includes over GBP 800 million of disposal proceeds in the year, fell to GBP 46 million. Reflecting this, net debt fell to GBP 2.6 billion. Return on average capital employed was 14%, above the 10%-12% boundary condition set out in our financial framework.
A net post-tax exceptional charge of GBP 476 million was recognized in 2017, predominantly relating to impairments of E&P assets and the Rough storage asset, but also including profits and losses on disposals from our divestments, and restructuring costs incurred as part of our group-wide cost efficiency program. Returning to adjusted operating profit, here you can see the split across our customer-facing divisions, with Centrica Consumer profit about flat compared to 2016, but significantly down in Centrica Business. Profit from our asset businesses was up, primarily reflecting increased production at Rough, following the change in its status from a storage asset to a producing gas field. Looking now in more detail at Centrica Consumer, where operating profit fell slightly to GBP 890 million. UK Home profit was up 1% to GBP 819 million.
Within this, energy supply profit was up 3% to GBP 572 million, with strong progress on cost efficiency and a focus on more valuable customer segments, more than offsetting a reduction in account holdings, warmer weather, and the impact of the prepaid tariff cap. U.K. Services operating profit declined by 4% to GBP 247 million, as strong cost efficiency delivery was not able to fully offset a 10% fall in gross margin from lower average customer holdings and increased pension costs. Ireland operating profit was at a similar level in both sterling and EUR, with the impact of competitive market conditions largely offset by cost efficiencies. North American Home profit increased to GBP 119 million, up 28% in sterling and 26% in USD terms, reflecting reduced losses from the closure of our residential solar business, a focus on more valuable customer segments, and cost efficiency measures.
Total cost per customer was down 2%. In Connected Home, gross revenue increased by 27% to GBP 42 million, reflecting growth in the volume of products sold and the diversity of customer offers. With incremental investment in the development of products, platforms, and apps, the business reported an increased operating loss of GBP 95 million. We expect 2017 to be the peak year of losses in Connected Home, as continued revenue and gross margin growth will outstrip operating cost investment. Overall, Centrica Consumer adjusted operating cash flow declined by 16% to just over GBP 1 billion, mainly reflecting the timing of working capital flows. Here's more detail on the drivers of year-on-year changes in adjusted operating profit in Centrica Consumer. External factors, specifically weather, foreign exchange movements, and the impact of the prepayment tariff cap, reduced operating profit by around GBP 134 million.
As I just mentioned, we saw an increased operating loss in Connected Home due to our choice to invest ahead of growth. Another like-for-like change reflects the loss of customer accounts. However, our efficiency program delivery of GBP 219 million largely offset these factors. Let me turn to Centrica Business, where full-year profit was down 67% to GBP 161 million. Adjusted operating cash flow fell 43%, reflecting working capital recovery in 2016 in UK Business, and materially lower operating profit in North American Business. Profit was down in each business unit, and I'll cover the drivers of the decline in the energy supply businesses, Distributed Energy and Power, and Energy Marketing & Trading in the next few slides.
In Central Power Generation, operating profit was down 53% to GBP 35 million, principally reflecting 2% lower generation volumes and a 4% fall in realized prices in our nuclear joint venture results. In a similar format to Centrica Consumer, here are the primary drivers of the year-on-year reduction in business operating profit. External factors, the combination of warmer weather, foreign exchange movements, and commodity price changes, had a small negative impact of GBP 8 million. Our choices to invest in Distributed Energy and Power growth and dispose of our wind farms and CCGTs reduced operating profit by GBP 35 million. However, we benefited from GBP 55 million of cost efficiency.
Nearly all of the change versus 2016 was driven by like-for-like decline, with poor performance in UK Business and North American Business in the second half of 2017, lower profits from our flexible legacy gas contracts in Energy Marketing & Trading, and lower output and realized prices already mentioned in nuclear. Let me now take each of these in turn. First, UK Business. UK Business profit fell in 2017 as the reduction in gross margin was not sufficient to offset lower operating costs from the efficiency programs and an improved bad debt charge, both of which were enabled by strong operational performance. The gross margin decline reflects a 9% fall in customer accounts due to high levels of market switching, lower unit gross margins from continuing competitive intensity on acquisition and renewal pricing. The effect of high wholesale electricity costs in the first quarter of the year.
Against this competitive backdrop, we are focusing our retention and acquisition activities on the higher value small and medium-sized enterprise segments with further development of our online offerings, greater emphasis on customer segmentation and value, and increased cross-sell of services and energy. As a result of these actions, we would expect to see a recovery in UK Business operating profit back towards 2016 levels in 2018. In North American Business, the operating profit decline was driven entirely by the fall in gross margin, slightly offset by lower operating costs. For the gas supply business, consumption was broadly flat, while warmer weather in the first and fourth quarters reduced the opportunity for wholesale optimization, and as a result, total gas unit margins declined slightly.
In electricity supply, we saw a 7% reduction in customer consumption, partly driven by a 3% fall in customer sites, but also the impact of increasing energy efficiency measures. In addition, unit gross margins fell by nearly 50%. Approximately half of this fall in unit gross margins was due to a GBP 76 million pre-tax one-off charge relating to a reassessment of a historic recognition of unbilled power revenues, with the other half reflecting a number of factors which I will describe shortly. As a result, total North American Business gross margin fell by just over GBP 200 million year-on-year, shown in the chart on the left-hand side of the slide. Total gas gross margin declined by about $40 million, reflecting the lower optimization opportunities I just mentioned. The biggest driver of reduced profitability in 2017 was the power supply business, shown in light blue on the chart.
This reflects the lower unit margins and volumes seen on the prior slide, as well as the one-off accounting charge. The North American Business power supply margins were significantly lower due to a combination of factors impacting the business at the same time. They included heightened competitive intensity, depressing acquisition and renewal margins, and changes to the market structure and related input costs, including higher unit capacity market charges. Realized margins were also impacted by reductions in customer volumes as commercial customers increased their take-up of energy efficiency measures, including distributed generation. Lower customer volumes also meant that unitized non-commodity costs were under-recovered. Finally, management actions in response to these pressures were hampered by poor visibility in our load forecasting and risk management systems due to the magnitude of internal change projects concurrently underway.
In response to these issues and the inherent risk exposure in power supply, we have undertaken a thorough review and investigation, and the business is implementing a number of changes. These include a new standard product offering that more closely matches input cost recovery, completion of system enhancements to provide greater granularity of gross margin drivers, and improvements to the processes and controls around load forecasting and risk management, and improving the level of capability. The aim of these actions is not only to improve the profitability of the power business, but also to reduce its volatility. As can be seen on the left-hand side of the slide, the performance of the gas business, with its physical asset positions and contract flexibility, has meant that gross margin delivery, even in years of extreme weather and volatility like 2014 during the polar vortex, has remained within a narrow range.
The power business is more affected by extreme weather and the knock-on impact that has to electricity price spikes and ancillary charge increases, where, unlike gas, we don't have the assets and positions to optimize and mitigate adverse factors in the same way. As a result, we have seen more volatility in gross margin in recent years, and is why we are taking actions with respect to the products we sell and the risk management processes we have in place. Going forward, we expect unit power margins to remain at 2017 levels this year, and therefore, after adding back the one-time accounting adjustment, we anticipate limited growth in underlying operating profit in North American Business in 2018. Moving now to Distributed Energy and Power and Energy Marketing & Trading.
In Distributed Energy and Power, revenue and gross margin were both up in 2017, reflecting organic customer growth and a full year of results from ENER-G Cogen, which was acquired in May 2016. Although the segment operating loss increased to GBP 53 million, driven by planned incremental investment in growth, we are seeing good momentum on building the future order book and recurring revenue streams. We expect Distributed Energy and Power to deliver continued revenue and gross margin growth in 2018. Although unlike Connected Home, we will continue to make further investment to drive this growth, and therefore expect the current year operating loss to be similar to that in 2017. Energy Marketing & Trading reported a 35% reduction in operating profit, despite continued gross margin expansion from the core route-to-market trading and LNG activities.
The capabilities we have built up over the past three years, as well as those acquired with Neas Energy, leave us well-placed to deliver ongoing growth in these areas going forward. Neas in particular, is delivering well ahead of its investment case. However, as you can see from the chart on the right, the level of gross margin from our three flexible legacy gas contracts, which we have been optimizing for value since demerger, declined. This will be a continuing trend as the two most profitable of the legacy contracts end during 2018. The one remaining contract is expected to be loss-making this year based on the current commodity price inputs that make up the contract's commercial terms.
This will create a year-on-year headwind for Energy Marketing & Trading operating profit in aggregate this year, and we would expect 2018 operating profit to be no more than half the level of 2017. Moving on to E&P, which following the disposals of our Canada and Trinidad & Tobago assets and the formation of Spirit Energy, will be focused on our European gas operations. Overall production was down 14% to 61 million barrels of oil equivalent, principally reflecting the impact of the disposals. Production in Europe was down 5% as a result of our decision to undertake asset integrity works at Morecambe to help improve safety, operational efficiency, and underpin the residual life of the asset.
Production from the rest of the European portfolio was similar to the prior year, with gas from Cygnus, which came on stream last December, offsetting the impact of the natural decline in the rest of the portfolio. In 2018, we expect production from Spirit Energy to be in the range of 50-55 million barrels of oil equivalent. Gas and liquids achieved sales prices were up in the year, which offset the lower production volumes and therefore realizations were in line with 2016. Operating profit was also commensurate with 2016 as total cash lifting and other production costs and DD&A were higher on a per unit basis, but with lower overall volumes, which therefore broadly offset each other. In 2018, we expect Spirit Energy unit cash lifting and other production costs to be similar to Centrica's in 2017.
Adjusted operating cash flow fell 32% to GBP 448 million, reflecting higher decommissioning spend and the phasing of Norwegian tax payments between years. E&P was still cash flow positive, even excluding the impact of disposals. Centrica Storage reported an operating profit of GBP 17 million for 2017, compared to a loss in 2016. This reflects better well performance and stronger gas production volumes at Rough following its return to service and the decision to produce a portion of the cushion gas for safety reasons in the fourth quarter. Having received final consent in January 2018 to change Rough's status from a gas storage facility to a producing asset, we would expect it to produce 8 to 10 million barrels of oil equivalent of gas in 2018. Going forward, we will report both Spirit Energy and Centrica Storage in one E&P performance segment. Turning now to costs.
Total reported operating costs were down 7% in 2017. After adjusting for items such as depreciation and amortization, impairments, smart metering, and portfolio change to get to a like-for-like number, adjusted operating costs declined 5%, and after excluding growth investment, they were down 7%. This reduction reflects the progress made in our GBP 750 million cost efficiency program. Including controllable cost of goods sold, we delivered a further GBP 308 million of efficiencies in 2017. Foreign exchange movements impacted our 2016 baseline by GBP 83 million, while inflation added a further GBP 103 million. When also including other net savings, not part of our efficiency program, total like-for-like controllable costs were 3% lower in 2017 than they were in 2016.
The efficiency savings delivered are from a combination of the annualization of 2016 savings and new 2017 initiatives, including the transformation of our customer operations, the utilization of digital and technology capabilities to enhance customer service and reduce call volumes, and the creation of a more integrated field operations model to drive efficiency and further supply chain improvements. We also saw a continued reduction in our global functions costs as shared service operating models became more embedded and the procurement function continued to leverage the group's scale to reduce third-party costs. Viewed over the last two years, excluding the GBP 103 million of cost efficiency delivered in the second half of 2015, we have delivered just under GBP 700 million of efficiencies over the last two years.
When including the end 2017 run rate, which will deliver GBP 54 million of additional annualized savings in 2018, we have delivered on our GBP 750 million cost efficiency target. We have done so significantly ahead of our original 2020 timeframe. Iain will talk about the next phase in cost efficiency in his presentation later. Moving on to net investment. Total capital expenditure increased 12% to GBP 943 million with the acquisition of REstore and the development of merchant assets in Distributed Energy and Power more than offsetting a GBP 79 million reduction in E&P CapEx. To GBP 439 million, towards the bottom end of our targeted GBP 400 million-GBP 600 million range.
As I mentioned earlier, total net investment was only GBP 46 million, after taking into account GBP 819 million of disposal proceeds, predominantly relating to the sale of the Lincs wind farm, Langage and Humber CCGTs, as well as Canada and Trinidad and Tobago E&P assets. Turning to cash flow, the disposal proceeds were a material contributor towards the net cash inflow of GBP 939 million. As already referenced, EBITDA fell by GBP 223 million or 9%, which when combined with a return to a more normal level of working capital in UK Business, resulted in a decline in adjusted operating cash flow of GBP 617 million. Cash interest payments increased in 2017 in the absence of net realized foreign exchange movements and a one-off interest benefit relating to the GLID wind farm disposal in 2016. While a higher scrip take-up on the first half dividend payment resulted in lower cash dividends paid.
Other cash flows were down slightly, with lower exceptional payments. I wanted now to turn to the group's balance sheet. We have been very successful in reducing net debt over the past three years, although given the maturity profile of our debt, this has resulted in an inefficient cash to gross debt mix. Reflecting this, we have announced today a GBP 600 million-GBP 1.1 billion debt repurchase offer. Including the one bond due to mature in 2018, we forecast an aggregate reduction in gross debt of GBP 1 billion-GBP 1.5 billion by the end of the year. We currently expect this early repayment will result in a one-off exceptional interest charge and cash outflow of between GBP 80 million and GBP 140 million, and generate annual interest savings of GBP 25 million-GBP 35 million per annum, and GBP 250 million-GBP 400 million in total over the lifetime.
With the range, in both cases, depending on the amount of debt we repurchase. This will materially improve the efficiency of our balance sheet. Finally, I'll cover credit metrics. We said at the start of the year, we were targeting to be at or above the financial metrics currently required for our existing investment-grade credit ratings. As a reminder, our rating with Moody's is Baa1 with stable outlook and BBB+ with negative outlook with S&P. We estimate we have achieved the target financial metrics at the end of the year, although, of course, the final calculations of the 2017 numbers are yet to be confirmed by the rating agencies, which they consider alongside other qualitative and business risk factors in determining the overall rating. With that, let me hand it back to Iain.
Thank you, Jeff. As I outlined earlier, in this second part of the presentation, I will provide a strategic update, and in so doing, also hope to address these key questions on the subjects of strategy and performance. Let me begin with a summary of Centrica's purpose and strategy. Centrica is an energy and services company, and our purpose is to provide energy and services to satisfy the changing needs of our customers. This is the core of the company, and we've been supplying energy and services to customers since 1812. We are re-emphasizing and returning to that core. However, it's not 1812. The needs of our customers are indeed changing, and as we deliver for them, the propositions we offer must change, too.
The reason for these changes in customer needs arises from three fundamental trends, which have arisen significantly as a result of the response to climate change and advances in digital technology. First, the energy system is becoming decentralized as new renewable technologies are developed and become viable, with many deployed at smaller scale and nearer to the point of use. Second, customers have more and more choice of how they can obtain energy supply and the services they wish to see alongside it. This means that power, influence, and rent is shifting towards serving the customer. Third, digitization and technology developments are accelerating the changes, enabling much more sophisticated management of the distributed energy system and enabling customers to have more control over their energy, their services, and the management of them.
In response, as an energy and services company, Centrica's strategy is to deliver for the changing needs of our customers. We aim to deliver long-term shareholder value through returns and growth, be a trusted corporate citizen and employer of choice, and to become a 21st century energy and services company. Centrica is therefore directing more investment into the customer-facing businesses. Centrica is also becoming simpler. We now only have three divisions: Centrica Consumer, Centrica Business, and Exploration and Production. Each has a clear participation strategy and strategic framework. The vast majority, over 90%, of our gross revenue and gross margin generation is within the customer-facing divisions. In 2017, unit gross margins on revenue were over 20% in both the consumer division and exploration and production.
Although much lower than this in Centrica Business overall, Distributed Energy and Power and parts of Energy Marketing & Trading also exhibited high unit gross margins. The reason we've established group wide consumer and business divisions is because we found that customer needs are very similar globally and many are seeking more than simply energy supply. Indeed, pure energy supply is commoditizing, and energy use per unit GDP is falling. In response, we've built new capabilities and propositions in both divisions, and the divisional structure enables Centrica to be more scalable, replicable and efficient. There are significant opportunities for growth associated with adding new propositions to energy supply with high unit gross margins. Technology is increasingly important in our propositions, and our customers, whether consumers or business customers, are responding very favorably to new digital platform offerings and innovations.
The world of energy supply and services is on the move driven by the customer. We are responding accordingly. We have developed the capabilities to deliver and established clear strategic frameworks for each of consumer and business. This is the strategic framework for consumer. The anchor of our business has always been in energy supply and in-home installation and servicing, whether on demand or through protection plans and warranties. This remains the core of our business today. Through installing boilers, heating and cooling systems, their meters, thermostats and controllers, we have always been in the home energy management business. As boilers become more intelligent and home energy management becomes more digital, with data analytics providing new insights and opportunities for the customer, we have had to develop the ability to install and maintain new home energy management systems.
This led to the development of Hive, starting with the digital thermostat and the intelligent boiler, Boiler IQ. Hive is the next phase of evolution of home energy management and a direct extension of our in-home servicing capabilities. Hive home energy management requires a digital hub to be installed in the home, and our customers are asking for other home management applications which can be easily served from the same hub. These are in areas of home security, remote diagnostics and home automation. Simply stated, consumers want the propositions within our four services pillars and are willing to pay for them. We're also finding that many consumers value receiving these services from the same provider as their energy supply, either separately or as part of a bundled offer. As a result, growth in services, including Hive, is a natural extension of who we are and what we're good at.
Our customers want it, like it, and it reinforces and leverages the historic core offerings of the company. Similarly, in business, we find that in addition to the commodity offerings of energy supply and energy wholesale, business customers are wanting access to more distributed energy generation solutions. We are moving from building large central power generation plants to many more smaller distributed units. Our customers want combined heat and power units, solar arrays, the grid operators want distributed power systems and technology to assist with the optimization of local energy markets and microgrids. Along with this, customers want to be able to gain insight from their energy use to save money and improve their operations, including in preventative maintenance. Customers who have distributed energy assets want to be able to optimize them and are willing to pay for optimization services rather than do it themselves.
The business customer is therefore also on the move, and we are responding to their needs. They want more than just commodity energy supply and the new capabilities and propositions we have developed once again reinforce the core of the relationship. Finally, the returns in value-added services tend to be higher than in pure commodity energy supply. The big message therefore, whether in Centrica Consumer or Centrica Business, is that the new propositions and services we have developed are not a distraction or somehow unrelated to our legacy businesses, they are to the heart of what our legacy businesses now have to incorporate. They are to the heart of what important and valuable customer segments are demanding. The good news is that Centrica has developed the capabilities to offer them in a high-quality way.
We are seeing improvements in the customer experience, the growth in demand is feeding through to revenues and at attractive margins. I'd now like to summarize what has been delivered during the first phase of the strategic repositioning of Centrica. As of the end of 2017, we are on track with the objectives we laid out in 2015. We have already reduced the resources allocated to the asset businesses by GBP 1.5 billion through lowering our capital investment into exploration and production by GBP 300 million per annum, with divestments of over GBP 900 million towards the top of our target range. We have created a new exploration and production joint venture, Spirit Energy, which will allow Centrica to participate in a stronger and more sustainable business while limiting our exposure. We have sold Canadian E&P, our wind assets, large U.K. CCGTs, and resolved the future of Centrica Storage.
We have, in turn, so far invested approximately GBP 700 million into the customer-facing businesses as we improve service, develop new propositions, and build capability both organically and through targeted acquisitions in response to the customer needs I described earlier. We have delivered on our material cost efficiency target of GBP 750 million per annum, three years early. Finally, we've maintained financial discipline, and in each of the last three years, we've ensured organic sources and uses of cash flow were more than balanced. With the proceeds from the divestment program, we have strengthened the balance sheet so that net debt is now towards the bottom of our targeted end of 2017 range. Let me now turn to what we need to focus on in this next phase.
In the next phase to 2020, as we continue to focus the group more on the customer and energy and services, although there remain some portfolio issues to address, the main focus is on performance delivery and financial discipline. In performance delivery, there are four key priorities we must deliver on, and all at once. First, crucially, we must demonstrate that we can grow gross margin through our customer relationships. Without getting to this place, Centrica cannot grow cash flow. We can rely upon a significant contribution from cost efficiency in the near term, but this must eventually give way to a higher proportion of cash flow growth from gross margin. Second, while we're improving gross margin capture and growth, in the near term, we must therefore drive cost efficiency as hard as we can.
During 2018-2020, we must aim to get as close as we can to being the most efficient price setter in our chosen markets, consistent with our targeted competitive position and brand. This is the basis for the next phase of our cost efficiency program of GBP 500 million per annum. Much of this will be focused on the U.K. in both our business units and group functions. Third, we must improve the effectiveness with which we operate and go to market. We have established a scalable platform, but we must become more agile, collaborative, and joined up. Fourth, we must also work to secure the capabilities that we will need for 2020 and beyond. The energy and services world is changing rapidly, particularly in areas of digital and physical technology, and we must have the capability to respond and develop new products and services.
We must do all of this while continuing to deliver improvements in safety, compliance, and conduct, and in operational excellence across the company, starting with customer service. This is the performance agenda for 2018-2020. It will require a huge amount of focus, determination, and delivery. I'm excited by the progress that we're already making and the momentum we're beginning to see. I'd like to begin with cost efficiency, and we'll return to gross margin in a moment. This slide shows the GBP 750 million per annum efficiency program delivery from 2015-2017, then the additional GBP 500 million per annum target we've announced today, and which will be delivered between end of 2017 and end of 2020.
As you can see, including other cost reductions outside of the core program, our actions from 2015 have resulted in a reduction of our overall nominal controllable costs from £5 billion per annum to £4.5 billion, while eating adverse foreign exchange impacts and inflation and funding our growth. On operating costs alone, again, the nominal costs at the end of 2017 are below those of 2015. The additional efficiency target will allow us to build on this track record and should materially offset gross margin pressures while aiming to drive nominal costs down further. We said in 2015 that by 2020, our nominal operating costs would be below those of 2015, having absorbed inflation and funded our growth, and we are on track to deliver this. Where will the additional £500 million per annum of savings come from?
As in the period from 2015 to 2017, about two-thirds of the savings will be from OpEx, with about one-third in cost of goods. About 60% of the total, or £300 million per annum, will be directly within UK Home in both energy and services and its supporting group functions. We estimate that in UK Home energy supply, the targeted efficiencies will deliver an improvement per dual fuel customer of £20 per annum by 2020 relative to 2017. These efficiencies will improve the underlying profitability of UK Home Energy Supply and will increase resilience in the face of any default tariff cap. The new efficiency program will involve an additional cost to achieve of GBP 300 million-GBP 400 million, and I regret it will also involve an additional direct headcount reduction of 4,000.
This is in addition to the 5,500 direct headcount reduction to date, taking the total reductions to 9,500, or a quarter of Centrica's end 2014 workforce. Against this, we will have created over 2,000 new jobs as we grow new propositions and businesses. The new program will be focused on the following areas: meeting customers' desire to self-serve through digitization of customer journeys, improving field operations, supply chain effectiveness and efficiency, using technology to improve productivity, continued transformation of group functions, including finance and HR, IT system improvements, and further procurement and supply chain efficiencies. The cost efficiency program is the anchor which will underpin net margins and our competitive position while we enable top-line growth, which I will turn to now. This slide shows the revenue, gross margin, and gross margin as a percentage of revenue for both the consumer and business divisions.
In consumer, although revenue has been falling gradually, as has gross margin, unit gross margins have been steady to rising slightly and have been over 20% in each of the last three years. Further, as customer accounts have been falling, the gross margin per account has been stable to rising at GBP 104-GBP 110. This is as a result of our focus on customer segmentation and on value, not volume. We are not focusing on customer segments and channels with negative or zero gross margin, because these will almost certainly be loss-making, especially if there are no cross-sell or up-sell opportunities. In business, gross margin has also been falling gradually. Unit gross margins are much lower, at about 7%, whereas gross margin per account was around GBP 650 in 2015 and 2016.
In 2017, for the reasons discussed earlier, business unit gross margins fell materially because of the issues in energy supply. Business energy supply has high turnover, lower unit gross margins, and can be volatile as a result of the impacts of weather and other factors. We will need to focus on reducing volatility of unit margins and improving consistency of returns in this area. This next slide shows the number of customer accounts in consumer. We have seen a fall of 1.35 million accounts over 2017. The majority of this reduction, 1.1 million accounts, occurred in the U.K. and Ireland, as shown by the dark section in each bar.
Of the total fall across the division, 85%, or 1.17 million accounts in the yellow bar, were either the result of choices we've made to end channels or not to renew collective switch populations or customers switching away, but in very low-margin channels. The gross margin impact of these 1.17 million account losses was GBP 6 million. GBP 6 million. Therefore, these accounts were loss-making at the operating profit level. Given the nature of these customer segments and channels, I am convinced we could not have improved their economics. Losing these accounts was a major driver in the increase in gross margin per account I mentioned earlier.
The higher value impacts are shown by the green bars, where we've seen a net loss of 180,000 accounts, which is actually an improvement of 96,000 relative to the 12-month picture we showed to the end of June, as we saw growth in both British Gas services and Connected Home, more than offsetting losses from other channels. You can see we lost 195,000 prepayment meter customers in the U.K., added 373,000 accounts in Connected Home, and lost 358,000 other higher margin accounts. Of these higher margin accounts, 214,000 were in U.K. Energy, but we were successful in moving a net 700,000 SVT accounts onto fixed-term contracts during the year in line with our stated objectives to encourage customers off the standard variable tariff.
The total year-on-year gross margin impact within the core portfolio, represented by the green bars, was GBP 128 million, approximately half in energy supply and half in services. The prepayment meter cap impact represented most of the year-on-year energy gross margin erosion, and in services, it related to pension and environmental program costs. It's our goal to stabilize the net position in the core portfolio and then begin to grow it. The unit gross margins in our profitable consumer channels are typically above 20%. If we begin to net grow revenue from these accounts, we will begin to be able to grow gross margin in consumer.
Zooming in on UK Home and looking at the interaction between gross margin and cost efficiency, you can see that although gross margin has been falling over the last three years, our cost efficiency delivery has been keeping pace with the gross margin decline, so maintaining EBIT margins per account at about GBP 39. Our goal is to focus on value, not volume, at the gross margin level while driving costs down to underpin EBIT margins. This is our focus across the consumer portfolio until we get to a position from which we can then net grow the total number of accounts. The resilience of consumer was demonstrated in 2017 with adjusted operating profit only falling by 1% despite intense competitive and regulatory pressures.
The ability to maintain competitive pricing, healthy unit gross margin quality, and drive cost efficiency is crucial in the face of the potential default tariff cap in the U.K. Let me therefore turn to this major external uncertainty and how we plan to deal with it. As you know, the U.K. government published a draft bill on a proposed temporary default tariff cap in October. This draft bill has undergone pre-legislative scrutiny in front of the BEIS Select Committee. We are against this intervention because we believe it's based on fundamentally flawed analysis of consumer detriment and will have unintended consequences, including negatively impacting competition, customer choice, and average prices for consumers. We've laid out our proposals for improving the U.K. supply market without a cap in the 14-point plan, which we published in November.
Irrespective of whether a cap comes into effect or not, we will continue to push for all 14 points, including those which we encourage the government and Ofgem to implement. This includes leveling the playing field for market participants and removing energy policy costs from people's bills, something which is highly regressive. Whatever the outcome, we are implementing the seven points we committed to unilaterally. These will improve the market, but also lower Centrica's exposure to any proposed cap. By March the 31st, we will withdraw the standard variable tariff for new customers, introduce a new default tariff, and measures to make it harder to end up on a default tariff. We will introduce new attractive fixed-term propositions, including fixed price, online only, and bundled tariffs. British Gas Rewards will drive customer loyalty and are already reducing churn.
Finally, we will continue to drive cost efficiency to ensure we are competitive with healthy returns, even in a cap scenario. In the meantime, we will continue to press the government and regulator on wider market reform. What is the risk to Centrica of a price cap? To understand this, you need to look at our exposure, our competitive position, and the result of our own cost efficiency plans. Firstly, we had 4.3 million customers on SVT at the end of 2017. We are expecting to have reduced this to about three million by the end of 2018 as the measures I described a moment ago take effect. This will reduce our exposure to any cap.
Secondly, as a result of our own cost efficiencies to date, our current SVT is cheaper than 85% of the SVTs in the market and is GBP 41 below the average of large supplier SVTs. As a result, if we were to maintain this competitive position, any price cap will impact the majority of the market first before it impacts us. The impact on those with the highest of all SVT prices, including some of the smallest suppliers, will be even more significant. This GBP 41 per customer provides a competitive buffer against the cap. When we introduce our new default tariff to replace the SVT, which will happen by the 31st of March, we intend to continue to make it competitive relative to the market. Thirdly, our efficiency program will deliver an additional GBP 20 per customer by 2020.
In terms of financial projections, our forward plans assume declining underlying unit gross margins by 2020 and now also incorporate an assessment of a temporary default tariff price cap impact from early 2019. Clearly, depending on the level of the cap initially, it's possible that the reduction in gross margin from the cap may happen more quickly than our cost efficiency can keep up, in which case, relative to our plans, we could see some EBIT margin compression, particularly in 2019. However, we must remember that the government has also said that when setting a cap, Ofgem must have regard to incentivizing efficiency, enabling effective competition, maintaining incentives to switch, and that efficient operators are able to finance their activities. Although the select committee has concluded it might be hard to meet all of these conditions all at once, the intent is clear.
The committee also underscored the need to guarantee the temporary nature of this intervention. Recognizing that the formula for any potential cap is not yet known, given the steps we are already taking, our competitive position and efficiency potential, we believe that we can deliver a sustainable energy supply business in the U.K. with healthy returns under most conceivable scenarios. Staying with the U.K., I've included, once again, this breakdown of British Gas dual fuel bills with the data for 2017 added. You can see that other than wholesale energy costs, which on average fell in 2017, the largest elements of the bill are those of delivery to your home and environmental and social policy costs. Our profit margin after tax was GBP 59 on an average dual fuel bill, once again in the range of GBP 42-GBP 65, which has been the case since 2009.
Over the last 6 months, wholesale prices have on average been rising, as have projected policy costs. Very recently, Ofgem announced a GBP 57 increase in the average prepayment tariff cap to a level just below our current SVT. We are monitoring such cost increases carefully. Let me now briefly cover some of the other indicators of growth we're seeing within Centrica Consumer. In North America Home, we've been growing protection plans as we learn from the U.K. experience and have seen 18% growth year-on-year. Local Heroes started from nothing in January last year, and we now have 7,000 technicians signed up, and we completed 25,000 jobs in 2017. As you can see from the graph, this on-demand offer continues to accelerate and complements our own contract relationships through British Gas.
In British Gas, on a half-year basis, we've seen the number of accounts in-home services grow in the second half of 2017 for the first time since 2011. Finally, British Gas Rewards sign-ups have now reached 700,000. Rewards allows us to enrich the relationship and proposition for our loyal customers in combination with our other offers and has reduced customer churn by, on average, 1.4 percentage points. These growing propositions demonstrate that consumers want additional services beyond commodity energy, and unlike many competitors, we are in a position to fulfill all of these needs at scale. This is also true in Connected Home. You can see the momentum we've started to build. We have seen installed hubs grow by 71% during 2017 to 900,000.
As of last week, we were at 950,000 hubs, therefore are likely to hit our 1 million hub target very shortly, albeit one quarter late. We did exceed our target for one and a half million products sold, delivering a cumulative 1.63 million by the end of 2017, and we now stand at 1.8 million as of last week. Revenue increased by 27% in 2017, with unit gross margins remaining attractive. Connected Home continues to grow in all of our geographies, and our first international partnership with Eni in Italy will see its full commercial launch in April. We forecast that 2017 was the peak year for net cash investment in Connected Home, and in 2018, we're now targeting a doubling of revenue, 500,000 new customers, and over 1 million incremental product sales. We continue to target GBP 1 billion of revenue from Connected Home by 2022.
This slide shows our competitive position in Connected Home. The Internet of Things market has thousands of participants. As you can see, in terms of smart thermostats, we now rank fourth globally and are the market leader in the U.K. and Western Europe. On the right, we rank seventh globally in terms of integrated multifunction ecosystem deployment, and again, are the leader in the U.K. and number 2 in Western Europe. We are competitively well-positioned in Connected Home. Other third parties are exploring partnerships with Hive, and it materially strengthens our other core propositions. Let me now turn to the business division and the area in which our performance let us down in the second half of 2017, North American Business. As we have covered, the demands of business customers are similar in all markets, and this is very much the case in North America.
We have a material and established business in the world's largest energy market, serving 240,000 customers in 24 U.S. states and eight Canadian provinces. We are the second-largest retail energy supplier by market share in the United States. The market plays to our strengths. It is large, we operate a large customer book, and it requires sophisticated energy price risk management, and increasingly the offer of a broader set of energy risk management and other services. The propositions of Distributed Energy and Power are increasingly important for the business energy customer in the U.S. However, the U.S. energy market is a volatile one, with significant weather extremes and regional differences. The acquisition of Hess Energy Marketing materially strengthened our capability, particularly in natural gas. Our recent track record is generally a good one, particularly in natural gas supply and optimization. In power supply, the track record is more volatile.
The issues we had in the second half of 2017 predominantly relate to the power supply book. As discussed on the call following the trading update in November, in addition to the accounting restatement, the weak performance in the power supply book was initially caused by compressed unit margins due to competitive pressures and lower volumes from efficiency and distributed generation take-up. As Jeff covered earlier, changes in the market, including backwardated capacity curves, combined with unsophisticated legacy power products, further reduced realized margins in 2017. Visibility on the degradation of performance and forecasting quality were simultaneously impacted by a new IT system under installation, which ironically is designed to give greater granularity on elements of power gross margin. Market change was occurring at the same time as we were managing significant internal changes.
In summary, the methods, products, and processes had served us well in the past, but we failed to adapt to a changing situation which required more sophisticated products, systems, and processes. Faced with this, it was imperative to investigate comprehensively what went wrong and why, and we've made changes to ensure we manage the book differently going forward. Our response, which Jeff also described, is designed to reduce the volatility of the North American Business power book, increase transparency and visibility of cost component risk, and improve planning and forecasting. We are also making some changes to personnel to improve capability. North American Business is an important part of the group. Over 2015 and 2016, post-tax ROCE was 10%, but in 2017 was a third of that. Adjusting for the accounting error, it would have been 6%.
The recent volatility in results is a concern, but returns have been generally attractive and the business has the potential to grow. The customer base and the capability we have give us the ability to offer new propositions and services which will add gross margin and stickiness to customer relationships, and we have the potential to grow the customer base. North American Business therefore fits within Centrica's strategic framework and has the potential to deliver attractive returns and growth, but we need to demonstrate that more consistently in the period to 2020. Moving now to our growth node of Distributed Energy and Power. We have built good momentum in this business. Active customer sites increased by 22% over 2017. Gross revenue increased by 6% and by 34% on an underlying basis when reflecting disposals and discontinued activities.
We also saw a 26% increase in order book revenue with accelerated growth over the fourth quarter. We have also put in place enablers for future growth. We've enhanced our demand response optimization capability with the acquisition of REstore, which complements the energy insights and energy solutions capability acquired through Panoramic Power and ENER-G Cogen acquisitions. We have transformed the way we go to market, including increased sales capacity, development of new propositions, and bringing our products together under one brand, Centrica Business Solutions. These enablers will drive an acceleration in the growth rate. In 2018, we are targeting revenue growth of at least 50%, and like in Connected Home, we remain on track to achieve our 2022 target of GBP 1 billion of revenue. Before I return to the group as a whole, let me touch on Exploration and Production.
The E&P division now consists of two business units, Spirit Energy and Centrica Storage. The formation of Spirit Energy has created a stronger and more sustainable E&P business, bringing together two like-minded shareholders. Our goal is to further develop Spirit through additional consolidation or partnership. We would expect to have a lower ownership percentage in any larger entity while maintaining exposure to E&P, and we would wish to retain sufficient influence to shape the strategic direction of the business. We would also be prepared to reduce our shareholding in Spirit Energy to below 50% if the right opportunity came along. Ultimately, the formation of the new business has created optionality for both shareholders, and as we said at the time of the announcement, we also do not rule out the possibility of an IPO in the medium term. Centrica Storage is now also an E&P business.
We will look to create synergies between Centrica Storage and Spirit while continuing to explore the commercial optionality of the Easington terminal. Touching briefly now on securing the capabilities we will need for 2020 and beyond in a changing world for energy and services. The environment is fast-moving with an increasing focus on integrating new technology. We have built enhanced capabilities over the past three years, both organically and through bolt-on acquisition, with improved customer service, customer segmentation, propositions, digital platforms, and technology. The technologies we focused on are directed at specific customer needs. In bolt-on acquisitions, we have targeted top-quality competitive capabilities, and the initial results are very encouraging. We have also accessed key new talent, both through acquisitions and through attracting new skills and capabilities organically. We have a strong focus on bringing forward the next generation of leaders and are working hard on succession planning.
Before moving on to our financial framework and outlook for 2018 to 2020, let me cover how we're thinking about acquisitions and disposals over the period. In the last three months of 2017, the risk envelope for Centrica changed, and as a result, we will not be pursuing any major growth M&A. This is because of the uncertainty over the U.K. price cap and our desire to maintain our hard-fought balance sheet strength. We may, however, make small bolt-on acquisitions to build our customer-facing capability within the capital reinvestment limits of our financial framework. For example, a priority would be to build out our capability in Distributed Energy and Power in the United States. Spirit Energy is likely to see a second-step transaction, as I mentioned a minute ago. Finally, a word on our nuclear shareholding.
Subject to ensuring alignment with our partner and being very mindful of U.K. government sensitivities in this area, we would hope to divest of our shareholding in U.K. nuclear power by the end of 2020. Having highlighted many of the developments in our businesses and hopefully answered some key questions, let me summarize the conclusions on strategy before I move to future financial guidance. We have completed phase one of repositioning the company. Centrica is returning to our strengths of energy supply and services. We have strong positions in those core areas. However, that core is also moving as customers demand different things and our capabilities and propositions are moving with it. In the near term, the uncertainty around the company has increased, largely because of U.K. political and regulatory interventions, and we face challenges of increasing competitive intensity.
As a result of this, as we mitigate the performance issues in North America, our focus remains on performance delivery and financial discipline. We will focus on customer-led gross margin growth. There are some encouraging signs from our new propositions. We will drive efficiency hard. We have significantly increased our cost efficiency target to 2020. We will maintain capital discipline and a strong balance sheet. We will not pursue major growth M&A. Let me now turn to the group financial framework and provide some financial guidance going forward. Our financial framework has been updated to show a revised CapEx target for 2018 to 2020, reflecting continuing capital discipline. We are delivering on many aspects of our framework, although we have yet to demonstrate consistent growth in adjusted operating cash flow. As a result, we have yet to restart a progressive dividend.
However, we have delivered on our cost and capital discipline. ROCE is currently running well above threshold levels. Our financial framework remains valid over the medium term. However, today we're also providing some specific guidance over the 2018 to 2020 period. We will be targeting on average GBP 2.1 billion-GBP 2.3 billion per annum of adjusted operating cash flow from 2018 to 2020. While we believe this is probably also deliverable in each discrete year, taking into account the unknown formulation of a potential price cap, there remains a slight risk to being outside this range in 2019. Capital expenditure, including the consolidated total for Spirit Energy and any bolt-on acquisitions, will be limited to GBP 1.2 billion per annum. We would expect the outcome to be in the GBP 1 billion-GBP 1.2 billion range in any year.
In terms of the dividend, we expect to maintain the current dividend level out to 2020, subject to two conditions. Firstly, being able to generate adjusted operating cash flow on average within the targeted range. Secondly, to manage net debt to within the range of GBP 2.25 billion-GBP 3.25 billion. A range designed to take into account a number of scenarios, commodity price projections. Consistent with maintaining our strong investment-grade credit ratings. We believe we can manage within both these constraints under most scenarios based upon our current projections. Clearly there are always risks associated with extreme commodity price movements and extreme regulatory interventions. Finally, in line with the group financial framework, we would intend to restore a progressive dividend when, in addition to the criteria already mentioned, underlying cash flow growth capability is demonstrated.
Our medium-term plans continue to show underlying AOCF growth rates of 3%-5%. Let me now bring this all together expressed as sources and uses of cash. This chart shows sources and uses of cash flow for 2015 to 2017. What we currently expect the picture to look like from 2018 to 2020 on average. Future projections are based on commodity price curves which were prevailing at the beginning of 2018. We're targeting on average GBP 2.1 billion-GBP 2.3 billion per annum of adjusted operating cash flow. GBP 1 billion-GBP 1.2 billion per annum of CapEx. With our dividend obligations, interest payments, other commitments, including pension payments, we would expect sources and uses of cash flow to be broadly balanced. Certainly to allow the group to remain within the target net debt range. This is obviously before any disposal proceeds.
We therefore remain confident in the sustainability of the group in the current environment, despite the uncertainties we face. Let me now cover specific targets for 2018. Adjusted operating cash flow is targeted to be GBP 2.1 billion-GBP 2.3 billion. CapEx is targeted to be below GBP 1.1 billion, including Spirit Energy CapEx of about GBP 500 million. In line with the earlier guidance, the 2018 full year dividend is expected to be flat at GBP 0.12 per share. We will target delivery of GBP 200 million of efficiencies as we progress towards a total of GBP 1.25 billion per annum by 2020. As part of this, like-for-like headcount is targeted to reduce by 1,000. We expect net debt will remain within an in-year target range of GBP 2.5 billion-GBP 3 billion. All of these targets assume normal weather patterns, current forward commodity price curves, and the absence of major operational outages.
Now let me summarize. 2017 was an extremely challenging year for Centrica, both in terms of our own performance delivery in business energy supply in the second half of the year, but also given the levels of political and regulatory uncertainty and intervention in the U.K. Despite these pressures, we delivered on our 2017 targets. However, the shareholder experience was extremely disappointing. We are absolutely committed to restore shareholder value and to demonstrate how the strategy will deliver this over the medium term. I hope today we've succeeded in addressing the important questions raised by 2017. Centrica has a clear strategy as we reposition the company back towards the core of energy supply and services. Phase 1 of the repositioning of the company has been completed with many important milestones met.
The next phase to 2020 is all about performance, delivery, and financial discipline as we marry customer-led gross margin growth with continuing to drive material levels of cost efficiency while maintaining capital discipline and a strong balance sheet. It is this combination and our track record of efficiency and financial discipline to date, which gives us confidence to deal with the uncertainties posed by any default tariff cap in the U.K., and to indicate a stable dividend outlook within defined boundary conditions. It is undeniable that at this moment, Centrica faces a higher level of external uncertainty. It's therefore imperative that we focus on the things we can control to underpin our performance, and that is what we're doing. This is how we will rebuild shareholder value and confidence, and I am determined to demonstrate this through our actions, one step at a time. Thank you.
Mark Hodges and Mark Hanafin will now join Jeff and me on the stage. We look forward to taking your questions. As ever, our Chairman, Rick Haythornthwaite, is also in the audience, should you have any questions for the board. Thank you. As usual, if you can indicate your name and affiliation, keep the questions reasonably brief, we'll head around the room. In the middle here. Thank you.
Thank you very much. Yes, it's Sam Arie from UBS. I have two questions. I also just want to say thank you for the presentation. I think it was extremely helpful this morning. I particularly appreciate the extra detail I think you gave us on British Gas and on the tariff cap situation. That's actually where my first question is on the British Gas business. You mentioned that the average EBIT margin is just under GBP 40 per account, but you also said you're succeeding in moving the customer base away from default tariff. Can you tell us what the average margin per account is when you exclude default tariff customers? Because I think that'd be a helpful indicator of sort of where the business might be post tariff cap. Hopefully, that one's an easy one.
The second question is more about overall results, where you landed for 2017 and what we should expect for 2018. My read is that you've landed absolutely bang on consensus for 2017, but with EBIT a little bit lower than expected, but some positives that you mentioned in the tax line. Can you just talk to us about how that rolls forward to 2018? How much, if you like, of the EBIT negatives are one-off and don't recur next year? Then what sort of tax rate we should expect for next year. I think consensus at the minute implies a step up about 10% from 2017 to 2018. So we're just trying to check if that's still fair. Thank you.
Sam, thank you for that. Well, I think Mark Hodges should answer the first one, and Jeff the second. Makes my life easier.
Thanks. Morning, Sam. It doesn't matter how much we disclose, there's always more. When you say default tariff inside British Gas, we don't typically disclose the gross margin or the EBIT profit. It's obviously a highly sensitive number. The thing I can tell you is that, Iain mentioned that we'd moved 700,000 customers from SVT to fixed products. What we're seeing is that we can move people onto our fixed products at margins that are very, very similar to our SVT book. We're not seeing a big degradation as we move those customers. If you remember back to last year, when we contacted all of our SVT customers, we had something like half a million move. This year, 700,000. Iain's pointed to 1.3 million customers we'd like to move during 2018. We're working with our customers to make that change happen.
We're not seeing a big degradation in margin. Obviously, overall competitiveness will end up determining the overall level of profitability from the book. The other thing is that GBP 39, just to remind you, also includes the services margin. That wasn't just residential energy, that's also residential services. That's a book, as we pointed to in H2 2017, that we finally started to see some growth. Of course, the ambition is to continue that growth trajectory now as we go into 2018.
Jeff, the EBIT and tax.
Let me take both of those. In terms of 2017 and the EBIT, Sam, you're right. There were a handful of things that we saw post the trading update in December that affected the full-year profitability. Virtually all of those were one-off or we wouldn't expect to repeat. They included the fact that our Sizewell nuclear station had to come off generation in early December. It's now back on stream. We saw in the E&P business, the Forties pipeline issues, which backed in some of our production. That issue has now been resolved. We had some extreme weather in the last week to 10 days in North America that made the, in particular, the NAB business slightly worse than what we were projecting. We would expect all of those to normalize out. I think the second part of your question was around the effective tax rate.
I'd probably think about it going forward in the following way. We disclosed the effective tax rate for 2017 at 22%. In the preliminary announcement, we flagged a number of items that were in 2017 that are less recurring in nature. Things like the U.S. tax rate change, as well as the level of PRT credits and provision releases. Although we tend to have some of that in a year, not to the level we would expect in 2017 every year. The other kind of boundary we then had within there was we said if you stripped all of that out, although we expect not necessarily all of it would come out year by year, you'd end up with an upper range of around 40% of an effective tax rate.
We wouldn't expect the effective tax rate to be that high, but I think we would, with the business mix that we'll see in 2018, more profitability from the E&P segment. Both with the combined Spirit Energy as well as with more production from storage. We would see a higher tax rate just on business mix alone, and therefore, it'll be more towards that higher range than the 22%, or the higher end than the 22% we saw in 2017.
Thank you, Jeff. We've got one at the back, Fraser, and then we'll go to Mark Freshney.
Good morning. McLaren from Merrill Lynch. A couple of questions on the new efficiency savings, please, and then one on supply costs. First of all, you suggested that the change programs in North America meant that management's eye was off the ball. As you embark on another substantial efficiency program across the group, what makes you confident you'll be able to avoid problems as you remove another 4,000 jobs? Just to check if the GBP 20 per account number is after inflationary pressures, and would therefore be around GBP 120 million net in U.K. supply. Are you therefore saying that your base case is no EBIT margin erosion as a result of the cap by the time we get to 2020 versus today? Last question is about upward pressure on external energy supply costs. You've delivered respectable supply margins despite having the lowest prices in the market.
To what extent is this helped by favorable hedges? When will they roll off? To what extent would higher input costs impact on margins this year without a tariff rise?
Wow. Quite a lot in there, Fraser. Let me touch on a couple of them, and then I will ask Mark Hodges to talk to the GBP 20 per account and the hedging impact inside the margin. First of all, the new efficiency program, generally, as with any efficiency program, we've got to be extremely careful. One of the challenges for Centrica as we go through this six-year repositioning of the company is clearly change. You're right that change, both market change and internal change, collided in the North American Business unit last year. I do want to emphasize that about three-quarters of the impact was to do with the accounting issue and market changes. It wasn't like all of it was down to our own change issues, but it was a factor.
Answering your question, look, over the last three years, in general, we've managed to drive efficiency while strengthening the system of internal control, improving compliance, improving safety, and very importantly, improving customer service. We've seen complaints, if I take U.K. Home Energy as an example, improve between 20%-30% on average each year. We've ended up, I think, 50% down on end 2014 levels. Or up, rather. Down on complaints, but better. This just does show that in general, we're able to walk and chew gum, if you like. Are you highlighting a risk around change? Absolutely you are. Our focus is going to be on managing change really carefully. Mark Hodges has a significant change team inside the U.K. That is where a lot of the change will occur, and I'm confident that we can manage it. Is there always risk?
There is always risk associated with change. The second bit I'd like to answer is just on this, are you saying your base case is no erosion to EBIT margins associated with an unknown tariff cap? Of course, we're having to make some assumptions. I indicated in my remarks that we have incorporated an impact of a tariff cap, that would imply we do see some EBIT margin impact from a tariff cap. I especially highlighted the issue of, while we're driving costs down, if they get out of sync with the gross margin impact of the tariff cap, again, we don't know what the formula is, of course we would see an impression on our EBIT margins until our cost efficiency program caught up.
We can't give you, we're not prepared to give you a forecast for what the impact is going to be. I hope that some of the indications around our exposure to it, our competitive position, and what we're doing on costs actually improves our relative resilience significantly. The other final point I'd make, why is relative resilience important? Because the government and the regulators are going to have to consider the impact on the market as a whole, not just on British Gas. We are significantly competitively advantaged relative to the average. Some other companies that are actually loss-making are going to be impacted by the cap as well. I don't know yet, but my assertion is that must be a factor, along with all those other things that Ofgem have somehow got to balance.
Mark, on the cost per account element of all that, and then energy supply costs and hedging.
The cost per account on Iain's slide contributes to the GBP 500 million, and that's pre-inflation. There will be an inflationary impact on that GBP 20 over the three years. On the whole hedging strategy, obviously, we don't know yet the mechanics of the cap. We don't know how it will operate. Not just what level it will be set at, but how often will it be reviewed. We're looking, as I've already said, to shift more and more of our customers onto fixed-term deals. That's something that we're used to in the way that we hedge handling. I think part of the question was, is there some kind of big event where a whole bunch of hedges roll off? That's not a risk, particularly. We're thoughtful about the commodity that we're buying into a price cap scenario and where we think the customer numbers will be by product.
We've been working very hard on that. I don't see a big event from a hedge perspective in the U.K. energy supply book. I just would go back and stress, I know Iain mentioned it on the change. It's something I'm very, very conscious of is the amount of change we're going through in the business. Just to go back to that complaint statistics, that's a million complaints less over the last three years in U.K. Home Energy, whilst going through cost reductions, whilst going through the headcount reductions, whilst improving service more broadly. We are starting, we believe, to build a track record of being able to balance the risk with the execution and make sure the customer really does see the benefit.
Thanks, Fraser. Mark.
Hi, it's Mark Freshney from Credit Suisse. Two questions. Firstly, on the British Energy stake, believe there are a number of restrictions on who the potential buyer could be and in terms of their competency. Can you talk about the stages that you would need to go through to assess and how likely it is that you could divest that? What you would potentially do with the proceeds, because the amount of net debt that you're running is actually fairly low relative to the underlying cash flows for the business. My second question is on Hive. You increased the number of products sold during the year, I think, by 70%, but the revenues only went up by 30%. Can you talk about what that disconnect is there, please? Thank you.
Great questions. Thank you very much. Particularly the last one, which I'll give to Mark Hodges to answer. Seriously. Firstly, on nuclear, I'm going to ask Mark Hanafin in just a second to talk about the setup, if you like, and some of the sensitivities. Just on the proceeds point, obviously, let's get there first, we have to recognize, as you say, first of all, that the balance sheet is quite strong. I hope I've indicated, and happy to work that with you a bit more, that the source and uses of cash ought to be reasonably balanced over this period. We do have other things that we have to take into account. We're going to have to spend the money on the debt repurchase program. We've got to spend money on the cost to achieve.
We've got other obligations, like our pensioners, that we've got to think about, as well as wondering about reasonable ranges of commodity prices. That's the, in brief, why we've got that slightly expanded net debt range. You're right, the net debt-to-EBITDA ratio of this company is now in pretty good shape. While the balance sheet might be a customer for some of the cash, we've got to think about if we ever get any, which is with a bit Mark will answer. We've got to marry up all the different possible beneficiaries of cash. Obviously, I've mentioned the balance sheet. We've also got to think about our pension obligations. Then, of course, we've got to think about the growth of the business, and we've got to think about our shareholders.
What I can undertake is that if we sell the business, we will look at appropriately balancing all of those things. Mark, on the complexities of this.
Yep. Obviously, the identity of a buyer is a very critical consideration in any sale. The sensitivity is not really about competence. EDF is the nuclear license holder. They are the competent authority. They run these assets. A new buyer, we're not restricted in that sense in terms of nuclear competence. The issue is more around the acceptability of a new buyer to our partners, and also to the government.
Thank you, Mark. Now Mark Hodges on Hive product sales and revenue growth.
Yeah. As you said, Mark, 71% increase in install base, 27% increase in revenue year-on-year. I think four reasons for the disconnect. We are changing the channel mix. As we moved into 2017 versus 2016, more through our retail channels, less through something like a British Gas channel, which is a positive. We actually have 46 retail arrangements in the U.K. We're in something like 1,700 stores across the country. The channel mix has an impact. The U.S. entry. We entered the U.S. working through Direct Energy. We're an unknown brand and an unknown quantity. We have primed the pump in the U.S. in 2017 for further growth in 2018. That had an impact. I would say there's a kind of product mix aspect as we move forward.
We're seeing people entering the Connected Home space, some of them in lower value overall products. A hub with some lights or some plugs as opposed to the thermostat as the first purchase. What that puts us in a position to need to do next is to be obviously cross-sell and upsell to the existing base. I think today a lot of the focus has been rightly on growing the total numbers. That will continue, as the held base gets bigger, we need to get much better at our CRM, at cross-sell and at upsell to the existing customer base. That will start to help improve that revenue percentage growth in relation to the overall product and hub growth.
You can do the math yourself. Our growth rate needs to be pretty high between now and 2022 to hit GBP 1 billion. Actually, it's approaching the rates already that in terms of sales growth. We would expect, obviously, revenue growth to catch up with sales growth eventually. Ian Turner. Then Deepa. I'll keep going around. Sorry.
Yeah. It's Ian Turner from Exane. You've announced the sale today, or the intention to get out of British Energy. You've pulled back CapEx. You're eschewing further M&A, potentially. Is this now a company that's in sort of runoff in terms of its strategic ambition?
I don't think so at all. I think that as we move from a company that has been very asset heavy and CapEx heavy, to a company that's more oriented around the customer, the capital reinvestment intensity should go down and indeed is going down. Actually, we're also deploying significantly more, and we believe the right levels of resource in order to grow the company at an appropriate rate. Also, we've had to build a lot of capability in new propositions and business models, acquire capacity and capability in order to start to grow the business at scale. It takes time. You're also right that it, or inferred in what you're saying, that with some of the challenges we've got, including political and regulatory intervention, that we've ended up with 13% cash flow growth one year and sort of negative 13% the next year.
We haven't yet demonstrated that the model can grow. However, from a cash flow perspective, it's not shrinking. Indeed, if you look at the last three years and strip out big working capital recovery programs like in UK Business, we delivered about GBP 2.1 billion-GBP 2.3 billion of cash flow. We're indicating a consistent GBP 2.1 billion-GBP 2.3 billion cash flow in the next three years. Now, a negative interpretation of that is ex-growth. A positive interpretation of that is that must require a number of sources of new cash flow to overcome some of the decline. Clearly, we've got to demonstrate that gross margin into cash flow starts to overtake the proportion that's coming from cost efficiency. I don't believe in any way that this company's somehow ex-growth or not pursuing its strategy.
I believe that we can demonstrate that the company's strategy will both pay for all of our obligations and be able to maintain the current dividend level, obviously within bounds that we've laid out. On your point about the major M&A or the inference there, you're right that we signaled last year that we were looking at whether to grow through acquisition. We didn't say how large. In the current environment, given the uncertainty and given the desire to maintain a strong balance sheet, we are not pursuing major M&A. Actually, the more important thing is we've looked at an awful lot of possibilities, and really there is nothing that attractive out there that we should be pursuing right now. Therefore, we can make that commitment, but we are going to be pursuing small bolt-on acquisitions, as we have been doing, to build out capability.
The simple answer to your question is no, I don't think it does. Deepa.
Thank you. Deepa Venkateswaran from Bernstein. I have three questions. Firstly, with Centrica Consumer, can you quantify what is the headwind from the weather impact in 2017? The second question is the transition of the SVT customers to fix that during 2018. What do you think is the impact on the gross margin from the 4.2 going to 3 million? Secondly, would there be other additional OpEx costs? I don't know, more extra staffing in the call centers to contact all these customers? If you could just quantify those two. The third one is on upstream. Could you clarify whether the unit lifting cost, you expect that to be flat at GBP 14.9 per barrel going forward? Thank you.
Let me take the lifting cost point and just say that obviously what tends to happen with costs in E&P is that costs follow price. The E&P industry has a habit of industry costs following price with a delay of about 18 months to two and a half years. That track record has been borne out recently. However, one of the main drivers of our lifting cost increase last year was actually the loss of production, i.e., the denominator, because we deliberately shut down Morecambe for a very long time in order to improve its reliability and safety performance. The increase is exaggerated in 2017. There are some inflationary pressures coming through in the industry, but not dramatically as yet. If prices stay high, I would expect inflationary pressures to continue.
On the weather impact in 2017, I mean, Jeff, we don't disclose the weather impact as a separate item, typically. We can say that across the whole company, depending on the metric, or whether it's pre or post-tax, GBP 100 million-GBP 200 million, I think, was the overall impact. Is that right, Jeff?
Well, we talked about in one of my slides that the impact of foreign exchange and weather and those factors was about GBP 130 million. Weather is less than that, and obviously some of it's in North America and some of it's in the U.K. I think as an estimate, you'd take half to two-thirds of that GBP 100 million.
It was the largest single factor in that. Mark Hodges moving the customer book.
Yes. Back to the SVT to fix. As I said earlier, we haven't seen a significant degradation in gross margin through the moves that we'd make. Obviously, at the end of the day, the gross margin is partially just a function of the market price. It's very hard for me to give a prediction because who knows where market prices will be going through the next 12 months. Our belief is that we can make that move and retain valuable relationships with our customers. I would also say segmentation is a big part of this. We've been stressing now, as you saw again today, value versus volume.
There are differences between groups of customers, depending on a number of different factors that we need to assess, whether that be their propensity to churn, their propensity to buy another product, their consumption, their debt profiling, the amount of calls we take from them. There's a lot of science going in, as we've explained before, into the whole segmentation model. With competitive pricing, with the segmentation, hard to give a prediction, but our experience to date has been there hasn't been a significant impact on gross margin. I can be more definitive on the OpEx. We're accounting effectively for any increased costs that we need to bear to make those transfers happen successfully for our customers within the cost efficiency targets that we're setting overall. I don't see a big ramp-up of costs being needed. Lots of this can be done digitally.
Lots of this can be done with the great people that we already have. Any small increase will be more than offset by the efficiency targets we've set.
Thank you, Mark. Martin, and then John Musk has been very patient, and we'll go to Chris.
Thanks. Martin Brougham from Deutsche Bank. A couple of questions. One, on the debt buyback, could you just give us a little bit more details as to why you think that is the sort of overall NPV benefit? You're obviously buying roughly at market value, I guess, for the bonds. Is there something structurally that means that you have a higher value than what you're being forced to pay in the market for the debt? Then secondly, obviously Ofgem's looking at the idea of this reverse auction for groups of sticky customers and sort of doing a trial at the moment. Do you think that's just a non-starter for sort of customer concern reasons? Are you factoring in that as a possibility that if it's successful and turns out to be highly competitive, that Ofgem might try and roll that out? What's your perspective on that trial?
Thanks.
Martin, thanks. That's very easy. Jeff, and then Mark.
Yes, I think on the debt repurchase, I think two things are happening. Obviously, spreads have moved in our favor in the sense that the premium we'd expect to pay on buying back the gross debt is less than it would've been previously. Effectively, we're buying back and arbitraging our credit spread against that. That over time, then makes it quite valuable to us in terms of our reduction in interest cost versus what we have to pay up front in terms of that premium. Therefore, get a better mix of gross and debt and cash on the balance sheet.
Just picking up Ofgem. Yeah, very aware of the trial, and a number of the other ideas that Ofgem are exploring. We are always concerned about data privacy and consumer consent to these kinds of things. It's a trial for a reason. We don't know if it will work. We're not convinced. The challenge that Ofgem believe they're trying to tackle is this concept of a disengaged customer base. From our perspective, if we can start moving people from SVT, which is why we're closing it to new customers, and while we're engaging our whole customer base to try and move them onto an alternative product, that means we've engaged with them. That means they've made an active choice. That means they don't come under the remit of either the Default Tariff Cap or these kinds of interventions that may or may not happen from Ofgem.
The strategy we're pursuing is the right one for a number of reasons. That's another good example of why engaging our customer base directly and moving them onto products that they're actively choosing is the right thing to do. I don't know how successful that will be. We have our doubts. As I say, customer privacy and customer data privacy is very top of our mind as these kinds of things are being considered.
Thank you.
Okay, thanks.
Sorry, was there anything else, Martin? Go on.
Yeah. Morning. It's John Musk from RBC. Two questions as well. Firstly, on the dividend, and don't want to simplify it too much, but, are you saying that 2018 dividends, you can be very confident of hitting your targets for the GBP 0.12, but in 2019, the main variable's obviously going to be the price cap, and that's where there's some uncertainty? On services or the home services business, can you just give a bit more color on why you think the customer number started to grow again in the second half? Is that your actions, or is that a changing market dynamic? Within there, you've seen a fall in the margin, the gross margin, by about 10%. Other than pensions, is that a change in the market again or something we should be worried about?
Look, on the dividend. Clearly what we've said is we expect, which is quite a strong word, that subject to the cash flow generation and the staying in the net debt range that we would expect the dividend to remain at the current level. Clearly, our confidence is highest in the near term, and there's more uncertainty further out. From my perspective, I expect that that's something we're going to be able to do. We clearly need to deliver within those boundaries. I think your assumption is a good one, but as always, the final decision on dividend at each period is a matter for the board. That is what we're signaling. In terms of services, Mark?
Yeah. I can guarantee you it's been a lot of hard work, a lot of interventions. There are a number of things have happened on services, which were going on actually through the first half of last year and even the back half of 2016, to get that growth to start to happen. I've mentioned a number of times in meetings pricing sophistication. We've been investing in the quality of our pricing capability, for what effectively is an insurance product, so that we can more accurately risk-based price, which is both helpful from a sales perspective, but also from a retention perspective. We've been through a huge amount of retraining, in our contact centers, so that our people are better equipped to sell the product. We changed some of the product features. There's just an awful lot gone on.
In terms of the way we incentivize sales, that's another area that we've been focusing on. The market dynamics at the high level are pretty much unchanged. There is a broad shift to on-demand. That's why the Local Heroes capability is very important because that not only meets a customer need, it's also ultimately a potential sales channel for the full services product. Then on margins and profitability, you did highlight the single biggest factor, which is that gross margin's down. That's because the cost of our engineers is within the cost of goods. It's within the gross margin. Obviously the pension cost impact that we had last year gets run through that line. There are other things that we can do. We're still working on the efficiency in services as well, more broadly, productivity. The biggest impact on margin was pensions.
Thank you, Mark. Chris Lavett, and then
Good morning. It's Chris Lavett from J.P. Morgan. A follow-on from Deepa's question on the transition from SVT to fixed. With those customers, if price protection is introduced at a level which is lower than your fixed tariff customers are paying, will you offer that price protection to those customers as well? On E&P, your liquid sell price rose around 7%. In 2017, that's European liquids, but spot prices were up considerably more than that, 25%-30%. Can you give us some sense for the shape of the hedge book going into 2018 and 2019? That would be very handy. On smart meters, are you concerned at the rate of the smart meter rollout industry-wide?
Ofgem and BEIS do seem to be putting some emphasis on the smart meters being rolled out to contribute to getting rid of the cap in the future.
On smart meters, Mark Hodges will answer that. We've indicated as part of our 14-point plan, that we think there are aspects of the smart meter rollout that the government should review. I'll leave that one to Mark. On the liquids realizations in E&P, I'll leave that to Jeff. Mark Hodges can add anything to this. The tariff cap is currently directed at default tariffs. Why is it directed at default tariffs? Of the sense that if you've defaulted, you're not engaged. The default tariff, the cap is not applicable to non-default prices. It's not applicable to non-default energy components in bundled offerings, and we therefore believe we have freedom to offer energy as a component in a bundle or in other formulations other than on a default basis, at different prices to the cap.
Until we understand the specific formulation of a cap and the calculation and how long it's going to last, currently, the legislation being proposed is till 2020 with a possible extension to 2023. It's difficult to understand where all the different offerings are going to be relative to that, but we're very conscious of it, and I'll leave Mark to add any comments. Why don't we start with Mark? Add any comments to that, smart meter rollout issues, and we'll come back to Jeff on E&P.
Not much to add. We obviously want to try and, from our perspective, minimize the number of people who end up on the default tariff. That's one of the things that we've said in our commitments by making sure that at the end of any tariff agreement we have, they get lots of choice of other attractive offers. We will try and minimize the customer numbers. To answer your question bluntly, I don't think we can give a guarantee about where other fixed deals will be priced going forward. It will depend on the nature of the product. Is it bundled? Does it come with Rewards and the benefits that the Rewards programs bring? It's not just a very straightforward comparison. On smart meters, just to reiterate again, we are supportive of the whole smart meter rollout. We think it's good for consumers.
We continue to do our research. We see a reduction in consumption amongst the customers who have smart meters. It's about 3.5% compared to the average, it's saving real money and hassle. We do have concerns, hence the point Iain mentioned about the seven-point plan. The biggest challenge to the whole program, as I see it, ultimately, is customer demand. This is an opt-in scheme. We can't force people to take a smart meter. We can resolve eventually the technology issues. It's frustrating that the DCC has gone back again to October. We continue to play a very active role in trying to make sure that works and that happens. I believe it will work. I believe interoperability of SMETS1 meters will happen.
For me, the biggest challenge that we all need to face into is ultimately customer demand and what do we do with groups of customers who just may not want to have a smart meter. I don't think there's an easy answer to that. Of course, the challenge becomes as well over time, it becomes expensive to keep chasing the same people. We all have to bear in mind as well that there's a marginal benefit from chasing the last groups of customers. I think that's where the debate needs to be had.
Okay. We've got Dominic and-
Sorry, I should just do that.
Oh, sorry, yes. Sorry.
Prices. We've talked previously, in our asset businesses, E&P, Rough or indeed power generation, we typically follow broadly a two-year ratable hedging strategy for the majority of the volumes we produce or generate. You could use that as a rough rule of thumb. With that type of strategy, the much lower prices we saw 18 months or so ago are starting to come out of those hedges and are at, currently anyway, at higher prices. Probably only the Rough volumes don't quite follow that because having just switched to a producing asset, it's more open to the current forward curve than the other assets.
Thanks, Jeff. Sorry, I forgot that. Dominic. Then we had one over here. There's a third. There's three hands up. Are there any more that Fraser's got a final question? I just want to check because I think we've been going a long time. There are five people want, can we keep them quick? Then I think we are going to have to close things down. Dominic first, please.
Hi. Yes, it's Dominic Nash, Macquarie. A couple of very quick questions then on your cash flow numbers, your GBP 2.1 billion-GBP 2.3 billion out to 2020. First of all, presentationally, how does the adoption of Spirit affect that on your consolidation of their cash flows into those numbers? Are you going to be proportionally consolidating that, or will that be 100%? Does that mean the underlying number is actually lower than those headlines? Secondly, it's pretty clear to me, correct me if I'm wrong here, that the dividend policy is now extremely linked to cash. Does that mean that if cash and earnings diverge in any way, the number that is more important for us, for investors to look at your potential dividend is your cash flow cover rather than your earnings cover going forward?
Let me answer this, if I may, Jeff. I think, first of all, the GBP 2.1 or GBP 2.3 does include Spirit, 100% on a consolidated basis. I wouldn't think about Spirit as somehow not being underlying. It's part of the group's cash flows. If you do the math around the cash flows, if you take GBP 2.2 billion as a midpoint and GBP 1.1 billion as the midpoint of the capital spend, obviously what's left with the dividend spend and interest payments on a slightly reduced basis, and no pension obligations, you can see that the organic cash flows should just more than balance in that scenario. Just briefly, the flex around the capital and cash flow, depending on the exact scenario in each year, recognizing we've got working capital movements from time to time that affect adjusted operating cash flow.
That's the reason for the slightly expanded net debt range. On your last point, our dividend policy has always been explicitly linked to cash flow from 2015. We laid that out very explicitly, and we've laid that out again today very explicitly. I think you can conclude from that, the primary driver of our dividend policy is our ability to pay for the dividends. Obviously, broadly speaking, earnings and cash flow should follow each other. They both derive from EBITDA on different bases, it is all about cash flow, because you can buy things with cash flow.
To follow up on that. If your earnings fell below GBP 0.12 but your cash flow was fine, then you would be comfortable paying a GBP 0.12 dividend?
Look, my answer to that is, companies do have periods of negative dividend cover from earnings and still can consistently pay. We've given no conditionality around the dividend to do with earnings over the next three years. It is to do with operating cash flow and obviously the net debt range. However, we're not signaling either that we're expecting to go into negative territory on dividend cover from earnings. It's not part of our financial framework. Here and then over there.
Very quickly. Elchin Mammadov from Bloomberg Intelligence. The dollar fell about 10% in the past year. Could you please give us a bit of sensitivity in terms of your earnings to dollar changes going forward? You've talked a lot about the commodity prices, but maybe not so much about FX. The second question is about the slide 50, where you talk about your market position in terms of Hive versus the other market players. We see that Nest is up there, near the top. Are you comfortable being fourth or fifth player in the market? Are you expecting to bridge the gap going forward as you roll into new markets? Thank you.
Foreign exchange, Jeff, and then Mark on our competitive position in Connected Home.
Yeah. I think a couple of ways to look at foreign exchange. Obviously, the impact that's most easily seen in the financial statements is around the North American operations, if it's against the dollar. You can fairly easily do that with reported operating profit, by and large. To a much lesser extent, the Irish business. I think the piece that is harder to frankly estimate at any particular point in time, because it's involved with the E&P cost base, is that, particularly in Norway, a lot of the costs are effectively because of the NOK, effectively in underlying U.S. dollars. We do get this play of an impact on the cost base, which even if the pound is weakening, is helping us from a revenue perspective, but is then offset from an underlying cost perspective.
That impact is much more muted, but can at times have an impact on our result, but much harder to estimate.
Mark, are we going to beat Nest?
We already are in the U.K. Look, I think Nest obviously have a very strong position in the U.S. Our stronghold is the U.K. That gives us, I think, a really great platform into the rest of Europe, hence the Eni deal. I think from our perspective, really interesting that you have another very large energy supplier who wants to be in the Connected Home space, who doesn't want to start from scratch, who has a field force, who can do the read across to what we've done and think that we're a great partner because we understand their business and we can help them. I think across Europe, we have an amazingly strong position given the start point. In the U.S., we have great technology, we have the DE brand to leverage, but it is harder because it's a more competed market.
We're having to think very carefully about how we go to market, what channels we will sell through, how much money we will invest in marketing. We're talking to a number of really important players. We had a great relationship with Amazon here in the U.K., for example. That's a relationship I would love to expand in the U.S. There are ways for us to take Nest on in their home market, but I think the way I would think about it is the U.S. and Europe, we could definitely leverage all of the experience in the U.K. more easily in Europe. In the U.S., we're going to have to really prove ourselves from scratch.
Centrica, I think there's room for us and Nest in the world. The penetration at the moment is still reasonably low. There's an awful lot of people out there who want energy and these types of services. Last time I looked, Alphabet didn't offer energy. So I think we can find our own way competitively.
Hi. Thanks. It's Nick Ashworth with Morgan Stanley. Firstly, quickly just on EM&T. You talked about the legacy contracts rolling off. You showed the slide on 19, which looks at the core EM&T and then the legacy. Is the core EM&T a good place for us to be thinking about over the next few years? How does the one final legacy contract evolve from here? What needs to change to turn that from negative to positive, and how much of a drag is that? Secondly, just going back to the 14-point document that you put to government in November. What's their response been? Have you committed at all to moving away from the standard variable tariff by a certain date?
I guess, given the run rate moving from standard variable to another type of default tariff, it sounds like you could even get to zero standard variable tariffs by 2020. Is there some sort of internal point that you're looking towards?
Look, on that last one, I'll ask Mark Hanafin to talk to EM&T. We've had very continuous conversation with the government about their plans. I'm regularly in touch with the Secretary of State and his department. We indicated that our 14-point plan was going to be deep and comprehensive, and they were impressed by those factors. Some of the things we're proposing they do aren't easy, politically aren't easy. I know that the Treasury are intrigued by the revenue-neutral nature of moving policy costs out of bills into taxation. The problem is it's a political issue. It's called do you risk stirring up the hornet's nest with rates of income tax and so on. We've got the government's attention on this. On the other issue you're highlighting, the government recognize that we, in some ways, want the same outcome.
We want to get people off standard tariffs. We're disagreeing on how, that's been an open dialogue between me and the Secretary of State, we just disagree on how. You're right also that it just might be that we find that our exposure to standard variable tariffs makes the impact of a price cap on default tariffs much lower than people expected. Let's wait and see. On your timing question, we're going to withdraw the standard tariff for new customers by the 31st of March. We can't forcibly remove everyone from standard tariffs. We did ask the government to consider, and the regulator to consider doing that market-wide, but they've passed on the issue. We tend to find that governments don't act when the politics of the situation is difficult. Mark Hanafin, EM&T.
On the core EM&T business, we do see continued growth in that business. We have a very good capability there, we see growth around route to market, origination, LNG and gas and power trading. The legacy contracts, there are three of them left. They've been profitable in recent years. We've got two profitable ones rolling off next year, this year, sorry. We're left for a number of years with a loss-making contract. These go back a number of decades. They go back, in fact, to the time where there were no gas indices to price off. Therefore, the loss position on that remaining contract really depends on that basket of indices, which are some quite strange energy indices in there, versus the gas price that we can sell at.
To give you an indication of it, Jeff referred to some guidance around operating profit being about half this year versus last year in EM&T. I would say that most of that delta is related to that loss-making contract.
Thank you. We'll go to Fraser very briefly and Gus, then I'm going to wrap it up, because I think we've been going for a while.
Thank you. McLaren from Merrill Lynch again. Just a very quick one. You've spoken in the past about how the LNG contract that starts next year might be loss-making in the early years. With the recent rise in Asian LNG spreads, have you changed your view?
That's one for Mark Hanafin.
Yeah, look, let me give you a quick overview of it. It's 91 million MMBtus a year that we lift. There's a fixed liquefaction fee of $3. We can load it on our ships, and we can take it anywhere in the world. We're exposed to that spread. If you look at the 2020 prices coming on in Q4 of 2019, let's take 2020 as the first year, and you look at recent forward prices for 2020, and you assume the backstop of the U.K., just NBP versus Henry Hub, it would be contributing about $1.40 to that fixed liquefaction cost. Or if you like, a loss of $1.60, just in terms of a simple movement of cargo from North America to the U.K. However, since we signed the contract a number of years ago, we've been building up our LNG capability.
We've been out there, despite difficult market conditions, selling, marketing, and about half of the volume in terms of price exposure is mitigated by that marketing effort already. That gives you an overview. Now in terms of how we see supply-demand. I still think that we're seeing an oversupply in those early couple of years. It's interesting that we haven't seen that oversupply that was expected this winter. Hardly any cargos came to Europe. Some commentators, notably Shell, are pointing to that and saying, "Look, it's quite difficult to predict what demand is, and maybe that glut isn't there." We're still assuming it is, but others are being more positive in terms of their outlook on supply, demand.
Thank you, Mark. Gus.
Gus Fox-Strutt. Just if I might circle back to Ian's question about asset shedding. Mindful of the block of sale of CCGTs earlier this year, well, last year rather, including Langage. Is this a signal that within this strategy that you might contemplate getting out to central power generation altogether?
Well, look, we signaled that to some degree, in 2015, that we saw the trend moving to distributed systems, we are therefore investing in batteries, rapid response engines, peaking, as well as installations in customers' premises. I think this is a trend that's been proven even more intensely than when we announced it in 2015. The second thing is, I think it's also been proven that the trend isn't looking like it's heading back towards large central power generation, you only have to look at some of the subsidies that it requires in order to keep capacity going. As a general sense, we will be heading away from central generation, we will only have, if we sell nuclear, the Whitegate power station in Ireland. We will be replacing that portfolio to some degree with more distributed assets. Thank you.
Well, ladies and gentlemen, this has been a long session, I think it's very important that we covered everything that you felt you needed to hear. Just a few points in closing. The first thing is that our performance was weak last year, I hope that to a degree, you can now look through that with some confidence. The second thing is, the largest impact on our shareholder experience last year was actually not our performance, although it was probably the last straw, it was actually this uncertainty around politics and regulatory intervention in the U.K. in our largest market. I hope we've given you some sense of how we're thinking about it and how we're going to deal with it. The third thing to say is that we have got a clear strategy, the strategy is all the elements of it are interrelated.
The one thing that I hope we've also demonstrated is that customers actually want the components of the strategy together. Many customers do, and it's not as though we've got some satellite activities that aren't related to our core business. This is all about reinforcing and strengthening the core of energy supply and services. Fourthly, given that there's a lot going on that we can't control, our focus is on what we can control, and it's on performance delivery and financial discipline, and that is something we've demonstrated over the last three years. Finally, what gives me confidence about Centrica going forward, the first thing are the people in this company and the capability that we've got. The second are the assets and market positions that we start with from this point. The third is the strength of our cash flows.
The fourth is the strength of our balance sheet. If you like, the starting point and the covenant that we can offer is a strong one. We hope that we can demonstrate our performance going forward through our actions. We look forward to updating you, with the trading update, and then at the middle of the year. Thank you