Morning, everyone. My name's Kristina Kazarian, I'm the head of the Midstream and Refining Equity Research teams over here at CS. This morning, I have the pleasure of introducing Kim Dang, who's the CFO at Kinder Morgan. Kinder Morgan is one of the largest energy infrastructure companies in North America, with assets spanning most supply basins as well as demand hubs, particularly high leverage to the long-term natural gas demand growth story. With that, Kim, thanks so much for joining us from Vail today.
Thanks, Kristina. Happy to be here. Some good snow lately, too. At Kinder Morgan, we have an unparalleled asset footprint. We are one of the largest energy infrastructure companies in North America. We've got a leading position in all of our business units, we've got the largest natural gas transmission network in North America. We're the largest independent transporter of petroleum products. We're the largest transporter of CO2. We're the largest independent terminal operator. We have the only oil sands pipeline serving the West Coast of Canada. If you look at our structure, we have two publicly traded companies, KMI on your left-hand side, KML on the right-hand side. KMI is almost an $80 billion enterprise company. It's got an investment-grade balance sheet and owns and operates diversified midstream assets in the United States. It owns a 70% interest in KML.
KML houses our Canadian assets, including the Trans Mountain pipeline and the expansion project. We took KML public in May of last year to help us finance the Trans Mountain expansion. We created it and set it up to be a self-funding entity, meaning that it could fund on its own the CapEx associated with its projects, the biggest of which is Trans Mountain. To date, that has been the case, we expect that will continue to be the case. Another thing I point out on this slide is that management is aligned with investors. Management and the board of directors own a 14% share of KMI. Our CEO, Steve Kean, gets $1 a year in salary. He gets no bonus, and he gets no options. He is aligned with the investors. If you look at our strategy, our strategy hasn't changed.
We want to own stable, fee-based assets that are core to the energy infrastructure. We want to maintain a strong balance sheet. We are investment grade-rated. We have reduced debt by almost $6 billion since the third quarter of 2015, we are currently funding all of our investment needs with internally generated cash. We're focused on controlling our cost, we don't want to be penny-wise and pound-foolish, we're also very focused on operating safely. We track 36 different environmental health and safety criteria and statistics, in 33 of the 36, we are better than the industry average. We also track on-time compliance. We track almost 540,000 tasks that have what we expect people to do and the timeline that we expect them to do it so that we make sure that we are not missing on a reapplication for a permit.
Our on-time compliance last year was 99.9%. We seek to make attractive investments. Generally, we target around a 15% unlevered after-tax return. That will vary depending on the risk of the cash flows. Some assets we expect higher, some assets we will accept a slightly lower return. Since 1997, we've invested over $60 billion in capital, either through expansion CapEx or acquisitions, roughly evenly split between the two. We are focused on being transparent to our investors. We publish our annual budget every December. We announce it, in January, we take our investors through a line-by-line comparison versus the prior year. We also compare to it our quarterly results against our budget in each quarterly earnings call. I think it's interesting. As I've said, we've invested $60 billion in capital.
I think it's interesting to look at our recent performance on those investments given the industry backdrop over the last couple of years. We've completed projects between 2015 and 2017 of about $9 billion. On that $9 billion that we invested in products, natural gas, and terminals, we have achieved a six times investment multiple. That is just the CapEx that we invested divided by year two EBITDA. We use year two EBITDA because this is something that we look at internally in the company all the time, and we do have some projects where EBITDA ramps. Over the last three years, even if you looked at it at year one EBITDA, the investment multiple would be 6.1 times.
I think the reason that we've been able to achieve these results is, one, a significant amount of these investments, the contracts with our shippers and customers are take-or-pay. We're getting those revenues whether or not our shippers use the assets. Secondly, to the extent that we make an investment that is not backed by a take-or-pay contract, we are very conservative. If we made a gathering or a processing investment, we're very conservative about what volumes that we expect to achieve. Thirdly, we spend a lot of time tracking our CapEx that we invest and making sure that we're bringing our projects in on time and on budget.
We meet on a monthly basis to go through the projects, where they are, how they're performing, what we're seeing on cost, we take all that information and we feed it back into our estimating group so that we can make sure that as they're estimating new projects, they have the most current data. Those are a couple of reasons I think we have been very successful there. Looking in 2018, our budget guidance that we went through in January at our investor conference, we're expecting adjusted EBITDA to be just under $7.5 billion. That's about 4% growth from 2017. We're expecting distributable cash flow to be about $4.57 billion. That's 2% growth from 2017. DCF per share of $2.05. In 2017, we achieved $2 per share, that's 3% growth.
The dividend per share coming off $0.50, going to $0.80, That's consistent with the three years of dividend guidance that we've laid out. That's a 60% increase in 2018 over 2017. We're budgeting growth CapEx of about $2.2 billion, which is down year-over-year. Net debt to adjusted EBITDA, we're expecting to end at about 5.1 times, which is consistent with where we ended 2017. We would expect based on this to generate $568 million of discretionary free cash flow. The way I get there is I just take the distributable cash flow of the $4.567 billion, take off the expansion CapEx of $2.215 billion, take off the dividend at $0.80. That leaves $568 million in discretionary free cash flow.
We will use that excess cash flow, one, to invest in incremental projects that we identified during the year that are above and beyond what we have in the budget in the $2.2 billion, to make additional share repurchases or to pay down debt. We announced last year a $2 billion share repurchase program that we expected to start in 2018. We were actually able, based on our 2017 results, to start that share repurchase in the fourth quarter of last year. Between the fourth quarter of last year and what we've done in 2018, we have repurchased about $500 million worth of stock. It's about 27 million shares. Breaking down our earnings before DD&A by business segment. 93% of our EBITDA is generated from pipelines and terminals. Looking at the breakdown between the segments, 56% is coming from natural gas.
As Kristina mentioned, we are heavily weighted to the natural gas sector. If you look at the breakdown further in natural gas, about 80% of our natural gas segment comes from what I call large diameter pipes. Long live large diameter pipes. Think of them like the interstate highway system. A little under 20% is associated with gathering and processing, primarily in the Eagle Ford, in the Bakken, and in the Haynesville. Products pipeline is about 15%. About 60% of products pipeline is associated with refined products. That's moving gasoline, diesel, jet fuel. About 40% is associated with crude and NGLs. The 15% of the terminals breaks down about 80% liquids terminals. Liquids terminals which includes our Jones Act ships. The liquids terminals, typically we're storing gasoline, diesel, jet fuel. That's our predominant business within the liquids. Then we have the Jones Act tankers.
On the bulk side, what we're doing, the 20% that's bulk is we're transloading coal, pet coke , and steel for customers. CO2 is about 11% of our overall business. About 7% of that is related to oil production. About 4% is related to what we call our sales and transport business. We own CO2 source fields in Southwest Colorado. Some people refer to them as the worst performing natural gas fields because it's almost pure CO2. We produce the CO2 in Southwest Colorado. We have pipelines that take it down into the Permian Basin, where we sell that CO2 to third-party customers. That's what our sales and transport business is. Typically, on those assets, we have take-or-pay minimum commitments from our shippers.
On the 7% that is oil related, there we are taking a portion of the CO2 that we're bringing down the pipe and utilizing it in our own oil fields. We have two primary oil fields, SACROC and Yates. We hedge that production, which I'm going to show you in a minute. Kinder Morgan Canada, 3% of our overall business. That's the Trans Mountain pipeline that moves oil sands production to the West Coast of Canada and gets that production a world oil price. Looking at the breakdown of our contracts now. I broke it down for you by our business segments. This is how the underlying contracts work. About 96% of our cash flow is independent of commodity prices for our 2018 estimate. About 66% of this is fee-based take-or-pay. That means that what customers are doing is they're renting space.
They're renting space on our pipelines or they're renting space in our terminals. Just like when you rent office space, you have to pay for that space whether or not you use it. Okay? In the short term, it doesn't matter if the customers are using it. In the long term, we want them to use it because that impacts renewal. In the short term, no material impact from usage on our revenues. About 24% is fee-based, you could have some variability in the volume. On the fee base, the fees are, they're flat fee, fixed fees. They're not tied to commodity prices. The volumes aren't guaranteed. If you look at the breakdown of that 24%, about 10% of that is in our natural gas pipelines, and that's primarily on the gathering and processing side.
Generally, those volumes are secured by dedications of economically viable acreage in the Haynesville, in the Eagle Ford, in the Bakken. If you look on products is about 9% of the 24%. Here on products, what we're doing is we're connecting refineries to markets. Okay? In the U.S., we have the most efficient refining complexes in the world. We've got refiners that run at pretty high capacity rates that are feeding markets in the U.S. Petroleum product demand, as I'll show you in a minute, is fairly stable. It generally grows with demographic growth. There's not a lot of variability in the demand there. The volumes on that system are pretty stable and grow with demographic growth. On terminals, that's 4% of the 24%. On that, 88% of our fee-based revenues are associated with high utilization assets.
What I mean by that is, our liquids terminals, people are paying for the tanks, whether or not they use them. We have a very high utilization rate at those terminals. What's in this bucket for terminals is the ancillary fees we get when they move product around our terminals. If they're turning the tank multiple times, given that these have been high utilization for a very long period of time, those revenues are pretty secure. It also has requirements contracts for petcoke and steel. On the petcoke side, here, if the refineries are producing the petcoke, we are transloading it and moving it for them. Relatively stable business there. Steel is primarily associated with Nucor, where we have requirements contracts. If they're going to move it through their terminals, we're going to handle it.
About 6% of our cash flow is hedged, and that's primarily in our CO2 group. Here, you look at that. We have a very consistent hedging program. For 2018, we have about 70% of our volumes hedged. If you look at the volumes that are contained within the 6%, if you think about our volumes on CO2, because that's not guaranteed when we're producing the oil, this is all associated with the oil production because the S&T are generally in the take-or-pay bucket. The oil production has come within 1.4% of our budget over the last 10 years. Unlike some of the primary production, the CO2 production is tertiary production. There are years and years of studies on these fields, you just have a better idea of how these are going to perform. You don't have as much variability in their performance.
4% of our cash flow is commodity-based cash flow, as I mentioned. If you look at our sensitivity to oil prices in 2018, it is $7 million in DCF for every $1 change in the price of crude. Our budget assumes a crude price of $56.50, which is pretty close to where we are today. We have a $12 billion backlog of fee-based projects in terminals, pipelines, and associated facilities that are generally secured by long-term fee-based contracts with creditworthy counterparties. If you look at the 85% of the backlog that is in fee-based pipelines and terminals, that's the $10.2 billion that you see there on the screen. We expect that $10.2 billion will generate about $1.6 billion in adjusted EBITDA. That's a 6.5x investment multiple. Very attractive returns on our backlog of projects that will come online over the next five years.
If you look at the $1.3 billion that's in the oil and gas production on CO2, there, we target to earn greater than a 15% unlevered after-tax return on those investments. That's just because those investments have higher risk, we target to earn a higher return on those investments. Let me talk a little bit about the underlying fundamentals in our business that drive stability and growth. First, starting with the natural gas market. If you look at demand for natural gas, the two big drivers of demand for natural gas over the next several years are LNG exports and exports to Mexico. You can see between 2017 and 2018, LNG exports are expected to grow 1.5 BCF a day. If you look at it between 2017 and 2019, they're expected to grow 3.6 BCF a day.
Tremendous growth and demand for natural gas from LNG exports. Exports to Mexico from 2017 to 2018 expected to grow about 300 million cubic feet a day. From 2017 to 2019, as they bring on more of their infrastructure, over a BCF a day over those two years, more of the growth coming between 2018 and 2019. Also, natural gas used for power demand is supposed to increase. Now, if you look, it was down a little bit in 2017. That's a function of warmer winter and some switching to coal. Longer term, we expect that natural gas demand for power generation will increase as a result of power plants converting to gas. On the upper right-hand corner, you can see U.S. refined product demand. That's gasoline, diesel, jet fuel.
You can see, as I mentioned, very stable demand for those products with some modest volume growth. That's primarily a f ixed cost business. If you can drop that modest volume growth to the bottom line, you can get some attractive growth from those assets. The other thing that we get on the products pipeline is we get an inflation escalator every year. Every year, right now, the inflation escalator is a PPI Finished Goods plus 1.23%. For 2018, it'll be a little bit over 4.4%. Crude oil production. You can see crude oil production, the forecast for the next couple of years expected to grow both in the U.S. and in Canada. Nice underlying fundamentals for the businesses that we're in.
If you take a longer-term look at natural gas, Natural gas, obviously, as I've mentioned, is over 50% of our business. You can see longer-term natural gas demand. Over the next five years, the projections are that it will grow by 19 Bcf a day. Over 10 years, 26 Bcf a day. 24% and 33% growth respectively. That's driven, as I mentioned a couple of times, by LNG export demand, industrial demand, and exports to Mexico, with supply coming from the Utica and the Marcellus, from the Eagle Ford, and from the Permian, and also some coming from the Haynesville. We believe we're well positioned for long-term success. We've got world-class midstream assets. We've got fee-based cash flows that are secure and growing. We've been disciplined in how we allocate our capital. We have a high bar for new investment opportunities.
We have had attractive project execution. I showed you the 6x investment multiple that we've earned on the projects that we've placed in service between 2015 and 2017. We've got a strong financial position. We've got an investment-grade balance sheet. We've got a lot of liquidity, and we're funding all of our CapEx with internally generated cash flow. We've got an experienced management team, and we're very transparent with our investors. Therefore, we believe that we will continue to deliver value to our shareholders. We will use the excess cash flow that we have to invest in high return projects to delever our balance sheet. Our longer-term target is to be less than 5x debt to EBITDA and to return cash to our shareholders, either through dividends or from share repurchase. We've laid out dividend guidance over the next couple of years.
We expect we'll be growing our dividend, as I mentioned, 60% this year and 25% for the 2 years thereafter. If you look at the business risk that we have, obviously you've got regulatory risk for our rate cases. You've got permitting issues as we're putting new projects into service. We try to take those into account both in our cost estimates to the extent we think it's going to cost us more or in time to place those projects in service. I've mentioned the crude oil production volumes. You have risk on those volumes, although we've been very successful at predicting what those volumes would be in our budget. You've got commodity price risk on about 4% of our distributable cash flow. You can see the sensitivity is laid out. You have the potential for cost overruns and service delays on our project backlog.
Although again, historically, we've been very successful at bringing those projects in on time and on budget. We have some economically sensitive businesses. That's primarily in our terminals, largely associated with steel and coal. On our Canadian business, we own 70% of our Canadian business, those cash flows are subject to FX risk. Obviously, environmental and terrorism. Those are insurable, but on the environmental side, if those should happen, that certainly is a reputational risk. Although we work very hard, as I mentioned, to stay in front of that and make sure that we are safely and efficiently operating our assets. On interest rates, we float on about 28% of our debt. If you had a 100-basis point increase for the full year, that would mean on January one, interest rates increased 100 basis points.
The other way to think about it is if you had a 200-basis point move over the year, that would average 100 basis points. That's about $100 million exposure. When we do our budget, we factor in the forward curve. In October and November, around that timeframe, that's generally the curve that's reflected in our budget. Our budget does have an increase in interest rates factored into it. I'm not going to spend a lot of time on this slide. This is for you guys to determine, but I just lay out a few facts for you. We trade at a discount to our peers. If you look at it on a price to DCF, we trade at 8.8 times versus 12.1 times for the average of our peers. If you look at EV to EBITDA, 10.7 times versus 12.5 times.
As I mentioned, we've got very good dividend growth over the next couple of years, 36%, and we've got very nice coverage now at 2.6 times as we continue to fund all our investment needs with internally generated cash. We think that it's a compelling investment thesis. 60% dividend increase for 2018. We are expecting to further increase that dividend by 25% in 2019, again, another 25% in 2020. We've got $2 billion of expected share repurchase, which we've done to date about $500 million, repurchasing 27 million shares. With that, I'm happy to take any questions.
Sounds great. I see some hands in the crowd. I'll probably go first, Kim. Kim, Trans Mountain makes up a big portion of your growth backlog, and feels like even since the Analyst Day, we get in flooded with updates and announcement. Can you give us where the project stands right now and what kind of milestones we should be watching for?
Sure. There's a slide in our investment conference for those of you who want more details that lays out some of the criteria that we're looking for. Just to back up for a second. Trans Mountain is the largest project in our backlog. When we announced our 2018 budget guidance that we weren't going to go into full construction mode, that we're going to spend at a reduced level from a full throttle push towards completion until we had more clarity and were sure that we could complete what we started. What we mean by that is we need to have a high confidence level that we are not going to get stopped in construction. We need to see that we're going to get the permits that we need. We need to see that permitting authorities can't hold us up.
We need to see that we are going to be able to acquire the lands that we need. We need to see that our NEB and federal order in council will not be somehow revoked by a court decision. Those are the things that we're looking for. In order to accomplish some of those things, some of those things we can have input into and some control over, some things we can't. The court decisions, we would expect if they tracked on the same timeframe as the decisions that were rendered in Northern Gateway, that we would see court decisions by mid-year. On the permitting, we have done a number of things to try to ensure that we have more confidence about getting those permits. We made two filings, essentially with the NEB last year.
One with respect to Burnaby, who was taking an inordinate amount of time to issue our permits. What we said is, "NEB, you have federal preemption here. We think that we have fulfilled the permit requirements and the permits are not forthcoming. Would you please exercise your paramountcy?" They did. Then we also asked the NEB, "Would you please set up a generic process that has a fixed timeframe to it so that we know if this happens with another permitting authority, that we can come in and have a timely resolution of those permits?" They gave us most of what we wanted on that. They extended the timeline a little bit from what we proposed, but not anything that we can't live with.
Those are some of the things that we are doing to try to ensure that we can have a clear path once we know the outcome. We still need to get a clear path on the lands. We still need to see where the court decisions are. We still need to see some of these permitting agencies, see how quickly they're going to act, but we've had some positive developments recently.
Great. I'll do another one unless I missed a hand. At the Analyst Day, Steve Kean talked about $2 billion to $3 billion of potential projects a year that could replenish the growth backlog. Can you talk about maybe some of the opportunity set across your portfolio, maybe even on Cheyenne, Double H, how you're thinking about some of the opportunity sets in terms of conversions with those projects as well?
Sure. Obviously, more recently, one project that we've added to our backlog, obviously, is Gulf Coast Express, which is a natural gas pipeline to move natural gas from the Permian to the Gulf Coast to feed LNG demand, to feed exports to Mexico. Very nice project with long-term take-or-pay contracts. Those are the types of projects that we would like to add. That project is beneficial to us because it feeds volumes into our existing Texas intrastate system. We're looking at conversion project, as you mentioned, of some of our assets. Only going to do it if we can get the contract structure and the pricing to generate the returns that we would like.
One of those potential projects would be converting Double H from crude service into NGL service, and then converting our Cheyenne Plains and a small portion of our WIC natural gas pipelines, which are underutilized Rockies pipelines, to NGL service. There's a lot more to come on that one, but that is typical of the types of projects that we would look to add to our backlog.
Perfect. I think in the interest of time, I'll leave it there. Thanks so much for joining us today.