I thank the Cathaoirleach and members. I will address the subsidy because, like everything else in this project, even the make-up of a simple number can be quite complicated. The overall capped subsidy is €2.9 billion which includes VAT of €355 million. In net terms, the maximum cost that could go to third party operators and the bidder is in the order of €2.6 billion. The €355 million in VAT would not leave the Department but would go straight to the Revenue Commissioners; therefore, the amount of the subsidy paid outside the Department is up to €2.6 billion. Within that amount, €480 million has been set aside to cover unforeseen and potentially unavoidable eventualities, the incurring of which expenditure will have to be proved on a case by case basis by the bidder.
We will be working with the bidder during the next ten years in terms of passing and connecting homes. The seven-year deployment plan includes connecting homes and runs for years and years. The €480 million will not go outside the Department until the actual cost incurred is verified as a cost that was not foreseen. It is for 14 discrete cost activities which we have identified with the bidder and which it could not, with any reasonable degree of certainty, put into its bid because they are subject to future events in years two, three and four. If we did not agree a contingency fund, the bidder would probably have brought it into its bid price as a risk premium price. We have provided this €480 million contingency in a very controlled manner.
That brings us back down to the overall gross or the cost that is committed within the 25 years of €2.1 billion. Within that figure, €1.8 billion is committed up to year ten and the balance is thereafter. The difference between €1.8 billion and €2.1 billion is for the cost of connecting premises and for the rental and infrastructure. Again, the capital related cost beyond year ten is only paid once it has been incurred.
In response to the question as to where we can reduce costs and where do we plan to reduce cost, these are numbers in a spreadsheet at the moment. We hope they will be much less than that when we return in five or ten years' time. That is the expectation. The figure will not and cannot be higher than provided for, but we expect it will be much lower. Why do I believe that will be the case? I will break down the capital cost of the project into three categories. The first is the cost to pass premises and there are 540,000 premises to pass. The second is the cost to connect the premises. The third, which is included in the first two costs, is the cost of renting infrastructure. We have a very good handle on the cost of passing, as did all bidders. It will cost, give or take, €1 billion to actually physically pass all premises in the State during the next seven or eight years. This includes BCPs and all the material that one has to buy, be it brackets and poles, including new poles, and fibre cable. There is also a requirement for 144,000 km of cable, which I will address shortly. It may not be 144,000 km of cable because that is based on bringing fibre to every single home, which we know is not likely to be the case. However, we factored that cost of fibre cable into the bid as the worst case scenario.
The other major cost in the cost of passing is the optical line equipment that one would have back in the Eircom and Enet exchanges. The bidder has said publicly that its preferred partner is Nokia. If it is Nokia or another company, one has to pay for the equipment. We have estimated all those costs in our internal models and in the models of the bidder and the other bidders. In terms of the cost of connecting and passing premises, our internal model up to the final tender stage showed a 5% difference because it is the most clear path of costs. How much does it cost to upgrade the MANS for the equipment, the Eircom exchanges and bring fibre cable from the poles? The cost of upgrading and remediating the poles and ducts sits with OpenAir. That is within the rental agreement with OpenAir. If we pay €20 per pole, the cost of replacing that pole and maintaining it for the next 35 years sits with Eircom. That is the rental cost. As we know, the rental cost is around €900 million over 25 years. As to how we will look at that rental cost of €900 million, we do not do this because ComReg will look at that. ComReg is 100% responsible for whether we pay €900 million, €800 million or slightly more. If ComReg were to determine that the pole and duct prices should be slightly more, we have factored that into our contingency of €480 million within one of the 14 buckets to which I referred. Again we have provided for a capped amount and the bidder is on the hook for ComReg making a determination over the next 35 years that the regulated price should be more than is currently provided for. That risk sits with the bidder, except for the small amount of money within the €480 million that we have set aside for the eventuality that ComReg raises the price. If ComReg were to reduce the price, we would take back 100% of what was factored into the bid. We will only pay the actual regulated price on the day or month or in a given year. We are not in a position to determine what that price might be because it is an independent process run by ComReg. As we speak, the regulator is going through a detailed review of the pole and duct prices and other relevant prices to the market generally and it will presumably publish documents in the coming six months on what the pole price should be for Mbp in other areas of the country.
The third area, which is the biggest unknown, is the cost of connecting to broadband, in other words, how much it costs to get from the curtilage of every single road into every single premises and farm. For all the companies that were at the table for the past three years, that was the major debating point in every single dialogue and every modelling session with the technical staff and engineers. It is very difficult to estimate costs in this area so what we have accepted in the bid model is a cost of between €400 million and €700 million to connect the 540,000 homes. We have to assume the number that will be connected in the next 20 years. That price paid will be paid based on the actual costs. If it costs €300 million, we will only pay €300 million. The maximum factored into the cost of connection is under €700 million.
We will work on every eventuality, be it technology choices or doing it in the most effective way, depending on the premises and where it is located. The demographics and geographics of the locations of premises off the curtilage of every road are unique to each single premises. We have to decide whether to go underground or overground, whether there is existing infrastructure from Eircom or the ESB or whether it requires a wireless solution. Those decisions will be made area by area and bit by bit over the next seven years with the bidder, with our full oversight. The bidder has our full oversight month by month and quarter by quarter in terms of what it actually spends. Month by month and quarter by quarter, we will seek to reduce the maximum subsidy provided for to the least amount required. We stated in previous documents that in a very optimistic scenario, which is quite doable, we could see that subsidy fall by up to €1 billion. We could get a cheaper passing cost, although that is not likely to be the biggest saving, and there are potentially significant savings to be made on the connection costs. Maybe ComReg will reduce prices and the company may not need any of the €480 million we have set aside for weather events, inflation and other eventualities in years two, three, four and five. If, in the best case scenario, none of those events happens and they are all avoided or avoidable - we will look at other ways of mitigating having to dip into those buckets - we will not touch the €480 million. Those are the elements we will look at in the next number of months. We are constantly working on this to find out ways in which the engineers and accountants can come together to reduce the cost. Ultimately, however, we must achieve the objective of passing and connecting all of these premises. That is the answer to the first question.
In terms of other countries that do clawback, the system we are applying is the same as the systems applied across every member state in Europe. It is required under state aid guidelines. These guidelines, which the European Commission views as legally binding on member states, specifically set out that all members states must include clawbacks, wherever possible, in contracts because it is so difficult to predict the future in terms of the cost of passing and connecting, the cost of regulatory prices or the demand on the network. We have had many discussions with Broadband Delivery UK, BDUK, the authority that oversees all broadband delivery projects in the UK on behalf of the local authorities. BDUK works with the European Commission in the same way as the Department of Communications, Climate Action and Environment in Ireland works with the Commission. It negotiates relevant clawback provisions with bidders. Again, the percentage clawback achieved varies from case to case and country to country. In Ireland, we have achieved a 60% clawback on excess profits. We also have a 100% clawback on any cost savings on the majority of the costs the bidder incurs. For example, if the bidder gets cheaper fibre prices in the market, we will only pay the market price it achieves and we will not pay the bidder a subsidy in excess of that. If the bidder has assumed a cost of fibre of €100 per metre and it turns out to be €70 per metre, we pay €70 and the bidder does not get the additional €30 per metre. That 100% clawback is provided for on what is in the bid.
Clawback is not unique to Ireland but standard across France, Germany and the UK. In the most recent example in the UK, BT submitted a bid which took a very pessimistic view of the number of people who would use the network but it turned out within one or two years that demand increased from 20% to 40% overnight. BDUK clawed back £700 million from the figure initially agreed to be paid out in subsidy.
That is going on continually in every single authority in the UK. The most similar project which will have similar clawback provisions, the R100 scheme in Scotland, has not been concluded. It is going through similar processes to those we have undergone. It has been decided to go with three lots, which is how we started. A gap-funding model will be used. In terms of the governance, the contract is quite similar to ours. It follows a very similar template and many of the advisers we have used have been on board.
What we are doing is not uncommon. There are concession models in different local authorities in France, but contract and clawback provisions within the first 20 or 25 years are also relied upon very heavily. That is the standard form for any gap-funded or concession model.
I will touch on the ESB before Mr. Neary gets into it. He is the technical expert on the differences between using ESB networks, Eir networks, and the tower and mast networks that would be used should a wireless option be chosen. The initial engagement with an ESB network solution occurred when SIRO was at the table. We had more than 880 hours of dialogue with all bidders. Between mid-2016 and when SIRO pulled out in September or August 2017, we went through what it would mean to use these networks with many representatives from the ESB Networks team, including engineers and accountants. We had the benefit of a couple of hundred hours of talking through the challenges, benefits and all of those things. Mr. Neary was at all of those workshops. We have a very deep understanding of what it would mean to use the ESB network. ESB Networks gave us all of that detailed information over many presentations. SIRO then pulled out and, as part the contingency planning we carried out, details of which can be seen in our reports and documents from the latter part of 2018, we engaged with the ESB in meetings and went through things with it in the context of the challenges involved and what would be possible both within any contingency plan and the current plan. Again, it is absolutely open to the current bidder to make use of ESB networks where it is most cost-effective and practical to do so.
Another actor that is very important in this regard is the Commission for Regulation of Utilities, CRU, because it sets the prices for access to that network. We have engaged with the commission, as has ESB Networks. That engagement is ongoing and will continue. I will hand over to Mr. Neary to discuss the technology aspect.