All right. Let's kick off here with Kinder Morgan. We've got President Kimberly Dang here. Very appreciative of you coming along. For those of you who don't know Kinder, I'm sure you all do, they move about 40% of the natural gas in this country. They are the largest independent transporter of petroleum products, the largest transporter of CO2, and the largest independent terminal operator. It's a lot of largest there. Really looking forward to your comments. Kim, off to you.
Thank you. We thought we'd start today with an overview, very big picture. Can you give me control over the slides?
It should be. Got you. Hang on. Sorry.
Okay, there we go. This is a very high-level look at global energy demand. This is a 2018 IEA World Energy Outlook. You can see what it shows is that natural gas and petroleum demand grows for years to come. If you look on the right-hand side of the screen, you can see where that growth is coming from, primarily from the developing economies. India demand is projected to more than double during this timeframe. China is projected to be the world's largest consumer of oil and the largest importer of natural gas and oil. By 2030, we still have 650 million people that lack access to electricity. The population growth and the urbanization and economic development in these developing economies are creating a demand for these resources. Where is that supply going to come from?
A lot of it's going to come from the U.S. If you look, the U.S. is expected to account for over 50% of the increase in global supply from 2017 to 2025. Reserves in the U.S. have continued to increase. They basically have doubled from the level of reserves that we had 10 years ago. By 2025, they project that one-fifth of the world's every barrel of oil and one-fourth of every cubic meter of gas is going to come from the U.S. The opportunity here is to connect the U.S. growing supplies into the growing demand in the world markets. The good news is we're in a great position to do that. We have an unparalleled and irreplaceable footprint that we've built over two decades. We have the largest natural gas transmission network in America. We've got over 70,000 mi of pipeline.
We're the largest independent transporter of refined products. We transport over 1.7 million barrels a day. We're the largest independent terminal operator. We've got 157 terminals, and we're the largest transporter of CO2. Our strategy is pretty simple. We want to focus on stable fee-based assets that are core to the energy infrastructure. We want to be a safe and efficient operator. As we contract these assets, we focus on getting a multi-year term on these contracts and getting either take-or-pay or fee-based. If you look at our current portfolio right now, over 90% of our cash flows come from take-or-pay or fee-based assets. We're focused on maintaining financial flexibility. Our 2019 budgeted adjusted net debt to EBITDA is 4.5x . We had recent upgradings, we're rated mid-triple B. We've got sufficient liquidity. Essentially right now, we're undrawn on our revolver.
We're a very disciplined allocator of capital. We have very high target return thresholds. We're conservative in terms of when we model these what we're willing to assume on the terminal value. Our goal through all of this is to enhance shareholder value. We do that by investing in attractive projects, by providing a nice, attractive dividend that is growing, and with excess cash flow, repurchasing shares. We believe that KMI should be a core energy holding. If you look, we've got a $40 billion market capitalization. We are one of the 10 largest energy companies in the S&P 500. We've got investment-grade rated debt. We've got $7.8 billion in adjusted EBITDA. We've got 25% dividend growth in 2019, another 25% based on our expected dividend in 2020. We've got a $2 billion share buyback program, of which we've already executed about $525 million.
We generate enormous cash flow. If you look over the last three years, we have generated almost $10 billion of cash flow in excess of dividends. You can see that 2016, 2017, and 2018 when you look at the gray bars. For those of you who prefer a GAAP metric, this is not an apples-to-apples metric by any means, if you look at cash flow from operations over that same timeframe, we've generated approximately $10.5 billion of cash flow from operations. In 2019, we're increasing the dividend, you can see the red bar goes from 1.8 to 2.3. Yet still, we've got $2.7 billion of excess cash flow in order to deliver value to our shareholders. Our dividend coverage, based on the dollar dividend that we have communicated, is 2.2x , a healthy dividend coverage.
What do we do with the cash flow that we generate in excess of our dividend? Essentially, we have four choices: pay down debt, pay the dividend or increase the dividend, invest in capital projects, or repurchase shares. On the balance sheet, we have now achieved our long-term target of 4.5x . We're essentially done there. On the dividend, we've communicated what we expect the dividend to be in 2019 and 2020. We don't have any near-term decisions with respect to the dividend. For the next two years, it really comes down to capital projects versus share repurchase. Let me talk a minute about how we think about the trade-offs between those two. On the capital projects, we target a threshold well in excess of our cost of capital. Right now, we target around a 15% unlevered after-tax return, okay?
If we have projects that have take-or-pay contracts that are long-term in nature from creditworthy counterparties, we're going to obviously take something a little bit less than that. If it's a CO2 project where you have some commodity or volume risk, we're going to require a higher return on those projects. Just to put things in the ballpark, 15% unlevered after-tax on average. If you look at that return and you compare that return to what we could achieve on share repurchase, when you're repurchasing shares, that's a levered return, right? Because you're buying equity cash flows. You have to take the 15%, which is an unlevered on the projects, and convert that into a levered return. That levered return at 4.5x leverage, depending on how the cash flows fall out, is going to be 25% or better.
25%-30% on a levered basis. We think that that is significantly in excess of what we could achieve on share repurchase, that to the extent that we can find projects that meet those return thresholds, that that is going to get the priority for our cash flow. We don't require that the share repurchase return be equivalent to what we can earn on capital projects, because when you purchase shares in a company, you're purchasing diversified cash flows, versus a project has single project risk. We expect that the capital project will return more than share repurchase, but the gap is sufficiently wide based on the numbers that we've run, that we think capital projects get the priority at the return thresholds that we've targeted.
To the extent that we have cash flow in excess of those capital projects, then we will use that to repurchase shares. This is something that we constantly reevaluate. This isn't something that we set one day and we live with that for the next five years. We're constantly reevaluating that based on share prices and other assumptions that we use to run those numbers. Let me point out, some people have said, "Well, what if you can't find enough capital projects?" If we can't find enough capital projects, then we'll do share repurchase, and that will bring value to our shareholders. Right now, our stock is trading at a 9.8x EBITDA multiple. That is going to generate value to our shareholders. Not the same value that a project will at the returns that we target, but it will generate nice value.
I want to spend a minute and focus on natural gas and U.S. natural gas. That's important because over 50% of our cash flow comes from our natural gas segment. We move over 40% of the natural gas consumed in the United States. If you look at the right-hand side of this chart in terms of U.S. natural gas demand, it's expected to grow from about 90 BCF a day in 2018 to 119 BCF a day in 2030. 29 BCF a day growth in natural gas demand over that timeframe. The largest growth is coming from LNG exports, which is about 14 BCF a day, and then power and industrial load are the next two highest with four BCF each. Which basins is the supply going to come from to meet that natural gas demand?
If you look, there are four primary basins in the U.S. that are going to provide that incremental supply. The Marcellus Utica, the Permian, the Haynesville, and the Eagle Ford. If you look at the other U.S. basins, they're actually expected to decline over that timeframe. Really four key basins in the U.S. Three of those basins are in the Gulf Coast. A lot of the supply growth and a lot of the demand growth is concentrated on the Gulf Coast. You can see the LNG export demand is concentrated in the Gulf Coast, the industrial demand, the exports to Mexico. That works well for us because that's where we've got a significant asset position and a primarily unregulated position in that market area. Right now our backlog of projects is about $5.7 billion, with $3.9 billion of that being in the natural gas segment.
That's why I've focused on the natural gas segment here, because that's key to our future growth. If you look at the EBITDA multiple that we expect to achieve on the $3.9 billion of the backlog that's in natural gas, we expect to achieve a 5.4x EBITDA multiple. Beyond the backlog, I'll talk about that, some of the opportunities there in a minute, but we think year in and year out, we can invest between $2 billion and $3 billion in projects. One of the things that gives us confidence of that, and you can go and look at a slide from our investor conference, is over roughly the last 10 years, we've invested about $2.5 billion on average per year. I want to talk a minute about two of our largest projects.
Two of our largest projects are on the natural gas side, Gulf Coast Express and the Permian Highway. Both of those projects are to take natural gas supply out of the Permian to the Texas Gulf Coast. Each of those pipes is about two BCF a day. Gulf Coast Express comes online in 2019. Permian Highway comes online in 2020. In the next two years, we're going to have incremental four BCF a day of gas hitting our Texas intrastate system. That is going to provide opportunities beyond what we have in the backlog, because there will be bottlenecks there will be new industrial demand that we will need to move that natural gas around on our system. We think that'll drive over the long term, opportunities in the market area. Turning to the liquid side of the business.
The growth in the liquid side of the business is not as significant as the natural gas growth, we still have some nice growth on the liquid side. It's similar to what's happening in natural gas. You've got a lot of supply coming out of the U.S., and the demand's in the rest of the world. It also lends itself to an export opportunity. You've got a lot of growing crude production, as I talked about in the U.S. The other thing that you have is you've got some of the highest, the best refining capacity in the world sits on the U.S. Gulf Coast. That refining capacity is being expanded. You probably saw ExxonMobil just announced 250,000 barrel expansion to their refinery on the Gulf Coast.
The U.S. is going to be a significant player in meeting the world demand on the liquid side. Here too, on the refined product side, we've got a very significant position on the U.S. Gulf Coast. We've got the largest independent refined products terminal in the U.S. that sits in the Houston Ship Channel. We've got 43 million barrels of total capacity. Right now we handle about 15% of the U.S. exports of clean product. The connectivity that we have in this facility is really unmatched and largely irreplaceable at any reasonable cost. We've got significant inbound pipelines. We've got 20 inbound pipelines, 15 outbound pipelines. We've got cross channel pipelines that connect us to refineries, barge docks, ship docks. We've got a very significant position on the Houston Ship Channel. Now looking beyond the backlog. We talked about the $5.7 billion.
What opportunities are there beyond the backlog? Well, as I showed you've got the four growing basins. You've got the Marcellus Utica, you've got the Eagle Ford, you've got the Permian, and you've got the Haynesville. There'll be opportunities in those basins. In some of those basins, we have gathering positions that we will be able to continue to expand. There also may be opportunities for takeaway capacity. On the market side, as I talked to you about on the Texas Gulf Coast, we think there'll be opportunities there. Maybe opportunities on the storage side as the LNG demand really grows and kicks in. ICF estimates that there's approximately $800 billion of North American energy infrastructure investment required to support the expected growth through 2035.
Those are the things that help to give us confidence that we will continue to find attractive places to invest our capital beyond the $5.7 billion backlog. This is a slide that we call the Tale of Two Cities. On the left-hand side of your screen, we did a drill down of S&P 500 companies. We started with look at the S&P 500 companies that have net debt to EBITDA of less than 5x . That takes the 500 down to 412. Look at how many are investment grade. That takes it down to 298 remaining companies. Look at those that have significant size and scale. Market cap greater than $35 billion gets you down to 127. Look at the ones that have an EPS CAGR of greater than 15% for 2018 through 2020. That takes you to 21 companies.
If you look at those that have a dividend yield greater than 5%. That takes you down to two companies. If you look at one that has a greater than 20% dividend CAGR from 2018 to 2020, that leaves you one remaining company. However, on the other side, you look at our valuation metrics relative to the S&P. We trade on an EBITDA multiple at 9.8x versus the S&P at 11.3x. If you traded at the average multiple, obviously there is significant upside in our valuation. Our dividend yield is almost double that of the S&P 500. Just to summarize for you here, we have strategically positioned assets. Those assets are backed by 90% take-or-pay contracts or fee-based earnings. We have $7.8 billion of adjusted EBITDA. We have $5 billion of distributable cash flow.
We have a 25% increase in our dividend between 2019 and 2020. We have a highly aligned management team, and we have an active stock buyback program. Of which we have already repurchased about $525 million. With that, we will take questions, and just introduce with me today, Anthony Ashley, who is our Treasurer, and Peter Staples, who is a Director in our group.
All right. Thanks, Kim. That was great. Maybe I will just kick off the first one here. We just had our private equity panel next door, and one of the themes was the public to private divide and valuation. I think Kinder Morgan is largely viewed as sort of bellwether for midstream trade at 9.8x, clearly a discount. What do you think maybe closes that gap? You guys have done a lot over the last few years that has been shareholder friendly between deleveraging, simplifying capital return, et cetera. What else do you feel like you need to do to close that gap? At what point does going private, I guess again, your circumstance become part of the conversation?
When we went private in 2007, we were a much smaller company. The other thing is we have done a lot to repair the balance sheet, and going private probably involves levering up. I am not saying that it is impossible. It is a lot different picture now than what it was when we went private earlier in our history. I am not sure exactly what closes that gap. We have from time to time sold assets, and when we can get very attractive valuations for those assets, and that delivers value to our investors, we will do that. There are no sacred cows. That being said, it is very disruptive to go out there and market your assets. That is not something that we do lightly. When people come in with offers that are attractive, we absolutely will evaluate those.
If it makes sense for our shareholders, it is something that we will consider and execute. I think from our perspective, that's on the last slide here, is we can't change the market sentiment. That's not something that we can control. What we can control is the decisions that we make. I think over the last couple of years that we have made all the right decisions in terms of improving the balance sheet, in terms of investing in high return projects, in terms of the growth in the dividend and the share repurchases with the excess cash flow, and targeting very high return projects. I think those have been all the right decisions. The disposition of Trans Mountain, which was a large part of what helped us get to the 4.5x debt to EBITDA.
Those are all the right decisions, and we will continue to make the right decisions for this company, and stay focused on making money.
Great. Any questions? Okay.
Do you have any upstream exposure? If so, what is your strategic disposition to that?
Sure. Upstream, one of the things you saw on here is that we have 90% take-or-pay contracts and fee-based. On the take-or-pay contracts, you don't have a lot of upstream exposure. The upstream exposure really, because on take-or-pay, your volumes are fixed and your price is fixed. Really there, the exposure really comes on the credit side. We've got a slide that in our investor conference you can see, we've got very nice creditworthy customers for the most part. Not a lot of exposure on that. On the gathering assets, those are largely fee based. That's a significantly smaller piece of our business because about 66%, I think, of our business is take-or-pay. Then the fee based, I believe, Anthony, is about 20%.
You have on the 20%, there you don't have price risk because it's a fee, but you do have some volume risk. There, you could have some, to the extent that there are less volume showing up, you could have some exposure. That being said, the primary basins in which we operate, Bakken, Haynesville, are our two largest gathering positions. There we've got very nice economics, and economics that work at current prices. You don't have a lot of exposure there. That's how. Summarize. Oh, on the CO2 side?
Yeah.
About 7% of our business is in the CO2 oil and gas production business. What that is we have about 4% of our business is CO2 sales and transport. We own the source fields in southwest Colorado. We own pipelines that transport that CO2 down into the Permian Basin. For a big piece of that volume, we just sell it to third parties under long-term take-or-pay agreements. For some of that volume, we take it, and we have really two primary producing fields. We've got five fields, but two are the most significant ones, SACROC and Yates. There we inject the CO2, and we produce oil. There what you have is you've got some volume risk on the oil production side, and then you've got price risk.
If you look at the volume risk, we have been within 1% of our budget over the last 10 years in terms of predicting those volumes. Because this is a primary production, this is tertiary production. It's very different. People have been in these fields for many, many years. You've got a lot of geologic information on them. You're really just extending out the tail. We've largely been able to call the production within a very narrow band. Then on the price side, near term, we're largely hedged. You don't have a lot of price exposure in the current year, and then probably about 50% exposure two years out. Really what you have there is more medium to long-term price risk on that 7% of our business.
If you look in 2019, that sensitivity, and this includes the sensitivity across all our businesses, but most of the sensitivity comes from the CO2 segment. It's about $8 million in DCF per $1 per barrel. Now put that in context of the $5 billion of DCF that we generate.
Hey, Kim. On slide 10, you talk about capital allocation priorities. Balance sheet, dividend, capital project, share purchases. In your response to Spiro, you talk about all the asset divestitures you've done, whether it's TMX or SNG a few years ago, and several others I'm sure I'm missing. Can you talk about where acquisitions fall into that capital allocation priority? I mean, you guys have built an empire on acquisitions. I'm just curious where that falls. Does that compete for capital projects? Does that compete for buybacks? Also in response to the public-private arbitrage, do you see acquisitions even out there or
Yeah
Is it just too far apart?
I think a couple of things. One, they would fall into the bucket, in my mind, of the capital projects. Where we're going to require similar type returns that we would. Now, you don't have to build on an acquisition. Okay? You don't have that same risk. You have to put that in the context of the overall return threshold. I could see that requiring less than the 15% return because you don't have that construction risk associated with buying in-place cash flows. That being said, we're not going to bid acquisitions to an 8% return. In the current environment, a lot of times we're not going to be competitive on a one-off acquisition.
To the extent that that acquisition ties into an existing asset and therefore we can gain value from that acquisition that no one else could, then I could see where we could achieve our return thresholds. It's something, again, that we are open to. We constantly evaluate to make sure that we don't miss any opportunities. I think in the context of the current market, those are probably going to be few and far between.
Maybe just a question. Oh, you want to?
Can you just give me a little more detail on the update on PHP? Where does the project stand, and how confident are you on the October 2020?
Yeah. Both GCX and PHP are going as we would have expected and as we budgeted. There is a difference in building in Texas and building in a lot of other places in North America. I think we have the right legal structure to be able to complete projects in Texas. Things are going well.
You guys have made a very compelling case for a long time that stock is significantly undervalued. Could you sort of walk back through why you're not even more aggressive at buying it back if it's undervalued by 50%?
Yeah. Again, it goes back to evaluating the returns on projects relative to the returns on share repurchase. I think the share repurchase opportunity, as we look, I mean, that's very rough. That gives you an idea of where that range sits on a levered basis versus projects that are 25%-30% on a levered basis. Okay? To get to the high end of the share repurchase, you need to be willing to assume significant expansion on the multiple. That's not true to get to the 25%-30% on the capital projects. We have been able to find significant capital projects that meet our return thresholds. We have been focused on living largely within cash flow and keeping our balance sheet at the 4.5x.
The combination of all those factors means that while we've allocated some dollars to share repurchase, about $525 million, we've spent more money on capital projects.
I think we're out of time, I can't ask you how many shares Rich is going to buy today. We'll do it for next time. Please join me in thanking Kim and the team once again for being here.