Before I get going, I'd like to introduce a couple of my colleagues, Anthony Ashley, our Treasurer and our head of Investor Relations, and Peter Staples, our Director of Investor Relations. We've got an intimate group here, so I can go quickly, and we can open up for Q&A here towards the end. As Christine said, I'm David Michels. I'm CFO of Kinder Morgan. On the first page here, this illustrates our footprint across the U.S. and across North America pretty well. We are an energy infrastructure company. We own pipelines and storage facilities to move products from supply centers to demand regions. In fact, we're one of the largest energy infrastructure companies in North America. Our primary product that we move is natural gas. We own 70,000 miles of natural gas pipelines across the country.
We touch 40% of all the natural gas supply that is consumed in America on a daily basis. Over 60% of our cash flows are generated from our natural gas storage and transportation assets. However, as Christine mentioned, natural gas is just one of the products that we manage, handle, store, and transport. We're a leader in multiple other areas. One of those is refined products, so gasoline, diesel, and jet fuel. We're one of the largest independent transporter of refined products in North America. We also own 157 terminals, which handle many different types of products. One of the largest ones is also refined products. We're a transporter of carbon dioxide as well, one of the largest in North America. The map on the right illustrates that pretty well and I think gives you a nice demographic of where our pipelines reach.
As you can see, we really stretch to every corner of the United States, a little bit into Canada, a little bit into Mexico, but primarily, in the United States and across it. The red lines are natural gas assets. The black lines are refined product assets. You can see just how extensive our footprint is, and that's one of our key competitive strengths is a very extensive footprint, and that allows us to offer flexibility and optionality to our consumers, to our customers. That's a very valuable, very unique set of characteristics that we offer our customers. I'm going to start and end with some of the key characteristics that we think make us a very unique investment opportunity for investors. Admittedly, we're a little bit biased, but nonetheless, I'll go ahead and walk through these.
We think we're a great core energy infrastructure company to put in your portfolio. Greater than $40 billion in market cap, a large-scale investment-grade rated. We recently upgraded to mid-BBB by S&P and by Moody's. We've got a 5% current dividend yield with growth coming. We've declared that we expect to increase our dividend from 2019 to 2020 by 25%, moving from $1 per share to $1.25 per share. That's on top of the 5% current dividend, which is more than two times the S&P average today. We also have a $2 billion share buyback program. We're large investment grade, and we have a bias to returning value to our shareholders. Our strategy is pretty simple. It's been pretty consistent over the two decades since inception. We focus on assets that are stable and have fee-based characteristics with contractual backstopping. Core energy infrastructure.
We look to operate our assets safe and efficiently. As I said, we look for assets that have multi-year contracts that protect our investments. Currently, our cash flows are comprised of 90% take-or-pay or fee-based cash flows. We'll go into that in a little bit more detail upcoming. We look to maintain financial flexibility. We've got a low cost of capital. With the recent upgrades, that's becoming even lower. We have ample liquidity, and we're a simple C corp structure, easy to invest in, and significant liquidity in our name. We've got a disciplined capital allocation perspective. We have conservative assumptions that go into the models that we use when we're evaluating new commitments and new projects to invest in. We have high return thresholds well above our cost of capital.
Again, a way to help us protect the investment, protect the shareholders' money that we're putting to work. We're self-funding most of our capital program, that might be a unique thing to say in this environment. In our sector, that's pretty unique. Many of our colleagues in our sector and our peer companies rely on the capital markets to help invest and grow their business. We use our organic cash flow primarily to fund our growth capital projects. As I mentioned before, we're focused on enhancing shareholder value. We've got attractive projects that we invest in. We're growing our dividend. We have a healthy dividend yield already. We're looking to repurchase shares. We have repurchased about $525 million worth of shares. We're looking to do some more. Importantly, we're highly aligned with our investors.
Our management team owns 14% of the company, a company with a greater than $40 billion market cap, that's saying a lot. This strategy's been pretty consistent since inception of the company, it's been very successful. We think with this strategy, the best is yet to come. All right. Environmental, social, and governance. ESG matters have become more and more important to our investor base, it's become more and more of a focus. In many ways, we think we already do and have done a very nice job in this area. In a couple of other ways, we've adapted, we're implementing best practices in some areas. One of those areas is disclosure of some of the policies that we currently follow and have, but haven't done a great job disclosing those and communicating those to our investor base.
We're doing that, we acknowledge that that's an area that is best practice, we should continue to keep that practice up. In other areas, we've already been a strong performer, we're just doing a better job now communicating that to the investment base. Disclosure of our safety performance here on the bottom right-hand chart. This is something that we've been tracking and disclosing to our shareholders for almost 2 decades now. It's a long time. We track our safety performance on a monthly basis and update that safety performance, post it to our website so that everyone can see how we've done and how we do relative to the industry on those 31 metrics. Have done that for a long time.
Just by chance, it happens that it turns out that we've done a very nice job and typically outperform our industry average on the majority of those metrics. Importantly, another area that we've done a nice job on is methane emissions reductions. This one, I'd love to say that we're just good people who are doing that out of the goodness of our heart and to protect the environment. In reality, that's in a large part driven by the economic reality of methane is the primary product that we move through our pipes and through our storage facilities. When we release or lose some of that product, we lose value, and our customers lose value. It's in our best interest economically as well as environmentally to reduce methane emissions.
We've done that, and we've been a leader in this category for a long period of time. Okay. Moving on to cash flow generation. We generate a lot of cash flow, and that's what we focus on at Kinder Morgan. That's the primary performance metric we look at, is something called distributable cash flow. It's really kind of like a cash earnings equivalent. We replace depreciation with maintenance capital and cash taxes with book taxes. Excuse me, take out book taxes, replace it with cash taxes, we get more close to a cash earnings proxy. We've generated between $4.5 billion and $4.7 billion from 2016 through 2018 on that distributable cash flow metric. That's almost $14 billion in three years. We're also expect to generate about $5 billion on that DCF metric in 2019. After funding our dividend, which is the red portion of the bars.
From 2016 to 2018, we generated $10 billion of cash in excess of our dividends. $10 billion that we were able to put to good use funding growth projects. We put some of that to use on our balance sheet, and we also put some of that to use buying back shares. We generate a lot of cash flow. We're able to sustain our dividends, even the growing dividend that we have projected here and still have enough remaining to substantially fund our growth capital programs. We generate a lot of cash, and you may ask, how stable is that cash flow? I think this page gives you a good sense for just how stable the cash flow is. 66% or two-thirds of our cash flow for 2019 budget is take-or-pay.
That concept means it's revenue that we receive from our customers, regardless of whether or not they actually even use our facilities. They're paying us for the right to use. Whether or not they actually use it is irrelevant. They're contractually obligated to pay us. A very high-quality form of cash flow for us. That's been consistent. We have had a significant amount of take-or-pay cash flows in our revenue stream for a long period of time. An incremental 25% of our cash flow is from fee-based services, fee-based contractual arrangements with our customers, that means it's not contractually obligated if they don't use it, but if they do use it, they pay us a fee. Very much like a toll road concept here.
We have a very small component, the very small little slurs down there, the 0.4 and 0.4 of commodity exposed revenues that we generate. We hedge those where we can in order to mitigate the short-term volatility in commodity prices that may impact our business. Overall, 91% of our cash flows are generated from take-or-pay or fee-based business. A very stable set of cash flows. What do we do with the cash? We touched on this a little bit before, but our first priority is the balance sheet. We've spent a lot of time bringing our leverage down to our target level, and that's 4.5 times net debt to EBITDA. Our investment-grade rating is very important.
We worked hard to bring our leverage down, and we were rewarded by S&P and Moody's by being upgraded from low BBB to mid BBB. That very strong balance sheet is very important to us as well as liquidity. We have a $4.5 billion credit facility that goes largely undrawn throughout the course of the year. We've got a lot of liquidity available to us as well. Our dividend, we've discussed that a little bit. Our 2020 incremental dividend that we expect growing from $1.00 to $1.25. The bottom line is we have an institutional bias to pay out cash flow to our shareholders in a responsible way, and we think that we've struck that right balance with our recent bumps in the dividend. Next, we fund attractive return on growth projects.
Those growth projects have returns well in excess of our cost of capital, and we think we do a nice job protecting our investor shareholder capital in that way. We have a $2 billion buyback program. We've used about a quarter of it, but we still have a long way to go before it's fully depleted, and if we have cash in excess of our dividend and in excess of the growth capital program that we're funding, some of that could go to the share buyback program that we have in place. As I mentioned, we've achieved our balance sheet goal, our target leverage level of 4.5 times or around 4.5 times.
I think it's important to note that our stability, our scale, our size, our customer base, our business profile, the low-risk nature of our cash flows allow us to maintain a good level of leverage here, a little bit above some of the other industries that you might hear at this conference. We think it's appropriate, and our rating agencies think it's appropriate as well, and we're very comfortable with this level. As you can see, we've reduced our leverage from 2016 to 2018 from 5.3 to 4.5 times. We think that's a good target to have. Moving on to the fundamentals behind our business. Might surprise some of you, and we've heard a little bit of surprise talking to some investors that the energy demand globally is expected to grow for decades to come.
This is an IEA forecast. You can see through 2040, there's good growth in energy demand for This has renewables, natural gas, petroleum liquids, coal, and nuclear. If you cut off at the natural gas, the red bar there, you can see good increasing growth for those products over time. That was globally. That was a global demand page. You should see that globally demand's increasing. Where's the supply going to come from? Increasingly, the expectation is that that supply is going to come from the U.S., and here you could see the projections also from IEA showing the U.S. production of oil and gas almost doubling from 2010 to 2025 to become the primary supplier of oil and gas globally.
In fact, by 2025, the IEA projects that the U.S. will produce one out of every five barrels of oil and one out of every four cubic meters of gas produced worldwide. Quite significant growth, and a lot of that going to emerging markets and emerging economies. Drilling down to the U.S. natural gas market, because that is our primary product and the product that we're most focused on with regards to growth in the future. Here you can see multiple categories of growth. It's not just the exports and the exports of liquefied natural gas, but it's also our domestic demand for power industrial, which is really petrochemical growth. Exports to Mexico, who is increasingly looking for exports from the U.S. to meet their demand for natural gas, and others, green across the page here.
While the main driver is LNG, we feel comfortable that there's going to be growth across the board, and we touch so many of these different plays that we feel like we're going to benefit nicely filling up our existing systems that aren't completely full today and achieving additional growth projects to meet that additional demand. To illustrate just how significant, though, the LNG export growth is expected to be, you can see there the BCF per day growth is 14 BCF a day. 14 BCF a day. That's off of a 2018 total of 19. You can see just how substantial that market is expected to be, and most of that's going to come out of the Gulf Coast of the U.S., and we have a significant footprint on the Gulf Coast to help support that. Let's see.
One other point on this page probably worth making is on the power demand side. Might also be a little surprising to see power generation growth for natural gas increasing here. I think it's important to note that there are still a substantial amount of coal plants that are operating that will be environmentally and economically phased out, as we've seen occur in the recent past. There's still much more of that to happen. Also renewable sources. As renewable sources come on to the grid, system operators are incentivized to have natural gas backstops to support them should the wind not blow and the sun not shine during peak load demand hours. After covering the natural gas demand slide, it's not too surprising to see that the majority of our projects are focused on natural gas. $4.3 billion out of our $6.1 billion project backlog.
These are projects that we're committed to, that we will develop. In many cases, they're under construction currently. We have contractual backstops and commitments from customers to support these projects. 70% of that is in the natural gas space. Most of the backlog is generated from extensions off of our expansions of our existing footprint, which makes us much more competitive to meet the demand for supplying natural gas off of our network. Accordingly, the capital efficiency of those projects is very high. That's why we have such a favorable build multiple on this page. You can see 5.5x capital versus the EBITDA expected to be provided by those growth projects. A very attractive set of returns, a very strong build multiple. Recently, that's proved out. We've been able to achieve those build multiples.
They haven't just been our initial estimates, we haven't achieved them. Looking historically, if you look at the two bars on the left, the 6.1 and the 5.9, those are investment multiples. The gray is the initial estimate. The red was the actual occurred, or actual experienced build multiple of 5.9. We actually did a little bit better than we expected, and that's for all of our projects completed in 2015 through 2018. The green boxed bars represent just the natural gas component of that, we thought that was important for folks to see because that's where most of our growth is going to come from. That's where most of our investment's going to go into in the coming years.
To see that we've actually done quite well building out our natural gas pipelines and storage facilities in the last three years, we're confident we're going to be able to do that going forward. A couple of those projects that we have in our backlog, the Gulf Coast Express and Permian Highway projects. It's a little bit small, but I think we can touch the high points. Gulf Coast Express is a 2 BCF a day pipeline connecting the Permian Basin with the Gulf Coast, and it connects down to the southern portion of our natural gas system that runs up and down the Gulf Coast of Texas. The Permian Highway Pipeline and the Gulf Coast Express Pipeline is expected to be in service in October later this year. It's a $1.75 billion project.
The Permian Highway project is another Permian natural gas takeaway project connecting the Permian to the Gulf Coast. This one connects further north, that's that green dotted line a little bit north of the blue one. It's expected to be in service in October of 2020, coming a year later than Gulf Coast Express. Both of these are serving as a critical takeaway path for Permian gas production, which currently is really stranded gas. Those of you who watch the market, this is really demonstrated by the negative Waha basis that we've seen over the last few months. Natural gas in the Waha Basin in the Permian is actually trading at a negative. They'll pay you to take it away. Our pipelines are critically needed and couldn't get in service fast enough.
Currently, the Gulf Coast Express Pipeline is on budget and on target for an in-service in October 2019. Touching again on the LNG export markets, this just further illustrates the growing need for LNG. We are not necessarily interested in owning incremental liquefaction facilities directly, but instead supplying third-party liquefaction facilities. Others will build them, operate their facilities, and we will support their facilities by providing gas transportation and storage services to them. We have got contractual commitments already in hand to supply 5.7 BCF of a number of the projects that are in service, under construction, and FID today. Those have an average term associated with them of 19 years. We have got nice long-term contractual commitments on a take-or-pay basis on those contracts.
We think we are very well-positioned to take advantage of the next wave of projects that are coming to the market that have yet to commit themselves to supply services. Outside of natural gas, we are also well-positioned to take advantage of the NGL and refined product export fundamentals that are facing the country. Here you can see, for the whole world, demand of liquids per day growing. There you can see most of that is coming from emerging markets, China, India. 2.2 million barrels a day of incremental growth of these liquids. We expect we will play a part on the refined product side of that. Not so much on the crude oil side, but on the refined product side. We have a great footprint on the Gulf Coast with export capabilities.
Here you can see on the bottom right-hand side of the chart what we have experienced on our docks since the beginning of 2016. An 18% compound annual growth rate of volumes moving across our docks. We are already participating in that growing export need, and we expect we will have additional opportunities as that trend continues to emerge. Here is how. That blue strip right there represents the Houston Ship Channel, which is a very industrial part of East Houston. We have multiple facilities across it, but I think most importantly, to speak to the export capabilities, we have 12 barge docks and 11 ship docks on that channel that will allow us to participate in that growing export market. That was not by accident. We have spent a lot of money invested in our facilities over the years. Over nearly $2 billion invested in the Houston Ship Channel hub since 2010.
Incremental to the actual docks and the export capabilities that we have there. We actually have one of the largest terminals in all of the U.S. in that area in terms of storage capacity with 43 million barrels of total capacity. We have the docks to accommodate exports and the storage facility capacity to accommodate a lot of demand. All right, we did not have time to touch on all of these opportunities. We touched on LNG and Permian Basin, a couple of our LNG footprint opportunities. Think it is probably worthwhile to touch on the storage opportunity, which is top row, second to the left. As the LNG market continues to play out and additional gas comes to the Gulf Coast of America to be exported, as you saw on the slide earlier, by 2025, that could be up to close to 15 BCF a day.
As that gas reaches those markets, if those facilities are unavailable, whether they have unforeseen outages because of extreme temperatures in the summertime, or LNG tankers can't reach the docks because of weather or storms, or go through unscheduled maintenance themselves, we think that the opportunity for us to provide high deliverability, high withdrawal capability storage might be an interesting and quite significant opportunity for us to supply additional services to these LNG facilities. Think that'll play out over time as these facilities really come into play. Right now, we've only seen about four BCF a day get exported on a daily basis. As that continues to ramp up and continues to increase, we'll see some of those operational capabilities and some of the limitations around them potentially play out. Wrapping up here in the last couple of slides.
Once again, we think this is a compelling investment opportunity. Pretty attractive set of investment characteristics. In fact, if you just took some of the investment characteristics of our stock and our security and compared those to the S&P 500, very few would compete. We took a crack at drilling down on some of them. If you took our net debt to EBITDA of less than five times, we were 4.5 times. Narrowed it further to investment-grade companies only. The 500 S&P companies get narrowed down to 294. If you narrow it further by companies only with a market cap greater than $35, an EPS growth rate of greater than 10%, and a dividend yield of greater than 4%, you're down to four companies.
If you look at those with just a dividend growth rate of greater than 20% from 2018 to 2020, we are one of one. Statistics can tell you anything, but still, it's a pretty compelling set of characteristics at our company. All right, the final slide here. Bottom line, we've got stable cash flow. We have a lot of it. We have an attractive dividend, which is growing, an aligned management team, and an active stock buyback program. We're proud of our accomplishments, and we think we're poised for success over the long term. As I mentioned earlier, we think the best is yet to come. With that, I'll open it up to any questions you all might have. Go ahead.
Thank you. Could you help us understand the economics of a pipeline? When you say the build ratio six times or five times, does that mean the implied return is 5% and you deliver it up? Could you give us a sense of what the ROE of a mature pipeline, please?
Sure. For those pipelines, those are enterprise value multiples.
Okay.
The equivalent returns for something like that would be more in the high teens on an unlevered basis. If you look at that on a levered basis, it would be well into the 20% area. Now, that's on a new pipeline project where we're not subject to max tariff rate under the regulatory regime. I think it illustrates just how attractive some of those projects are. That's also why we look to secure long-term contracts. A couple of those pipelines we talked about were 10-year contracts on the Permian pipelines we talked about. That protects our investment over that 10-year period. It basically allows us to return the majority of our return on and return of capital during the contractual phase.
Can I ask also on the cash flow on page eight, did you have more commodity exposures five, six, seven years ago? Was it different? I'm referring to page eight. 96% of the cash flow is basically contract or stable.
How long ago?
Call it five, six years ago.
Oh, yeah. The reason I ask is we combined a number of our entities back in the end of 2014. From that point forward, we've been relatively consistent. Our commodity exposure has reduced during that time. It is now about 6%. Probably back then it was, I don't know, closer to 10%. We have seen it reduce. Prior to then, one of our entities that we had publicly traded had a larger exposure. After rolling them all together, it has been 10% or less ever since.
Okay. Maybe one last one for me, if you don't mind.
Sure.
In terms of the distributable cash flow, because you compare it to EBITDA, you don't have much maintenance CapEx. From EBITDA to distributable cash flow, how do you think about that? Because you seem to be doing everything at the same time. Are you paying dividend, having a lot of projects, buying back stocks? Is the answer the leverage or how come you can do all of that together?
I'm sorry, is your question what is the difference between EBITDA and DCF?
With 2024, maybe that will help.
Okay.
EBITDA $7.8 billion, distributable cash was $5 billion. Yet you have to pay taxes.
Right.
You have maintenance CapEx. You spend between $1 billion and $3 billion CapEx every year. That seems almost too big a number. You see what I mean?
Yeah, let me walk through it. I see where you're going now. Okay. To get from EBITDA to distributable cash flow, you're right. You've got our interest and maintenance capital and cash taxes. When we rolled our assets together back in 2014, it was a taxable transaction, and that allowed us to increase our taxable depreciable base. We have a number of tax assets that are generating a taxable loss for us, have for some years. We're not paying any material amount of federal income taxes, cash taxes on an annual basis and don't expect to for many years to come. That's one component of it. We do have maintenance CapEx. For 2019, that estimate is $715 million worth. Distributable cash flow is before funding our growth capital projects. I think those are the two pieces that might help bridge you.
Does that get you what you're looking for?
Yeah, I think it does.
Okay.
You were an MLP before, right?
We were.
You C corp, but you don't pay tax because of your tax assets.
That's right.
Maintenance CapEx $750. Okay. I think I get it.
Yeah. That's right. I think the big disconnect there is the cash tax is because of our tax assets that we have. There's one right back here.
Thanks. Yeah. This is another one on returns, actually. If you could just compare the incremental returns on capital for the new projects versus the sort of the base returns.
Right. It's a good question that's come up recently. I think if you look at the base returns today, they're much lower than what our growth capital project returns, that I've been talking about, are and have been. Over the last three years, you saw the multiples. The returns have been very good on those projects. On the base business, we've seen some headwinds, and we've got some information in here that have illustrated some of the headwinds that we faced on the base business. Typically, our base business is pretty stable. Since 2014, we've seen some headwinds from our commodity price-exposed assets reduce some of our EBITDA. Some midstream assets that have reduced volumes flowing through them. Now, that means that now they're underutilized, and we can see some uptick in those revenues associated with those assets because they're underutilized.
The other one was, we had a number of customers who were coal customers who went bankrupt, and we lost some revenue that way. I think those items put together meant our base business had some headwinds in it that we don't think are recurring headwinds. Actually, one of them might turn around now. That has really weighed on some of the returns that we're seeing in the overall business today and masking the real returns that we're generating on those growth projects.
Thanks. On the new projects, the construction risk, is that borne by you, or do you subcontract that out?
We have contractors for the majority of our projects, and we look to enter into contractual arrangements with our contractors where a number of identifiable risks are shared or are pushed onto the contractors where we can. We've done a nice job here recently of working with our contractors and sharing a reasonable amount of that risk. I think that's part of the reason why you saw this good performance on our projects in the last three years is because of that arrangement that we're looking to achieve.
A final one from me. Just on the new contracts, is there a shift towards the take-or-pay versus the volume fee-based? Is there a discernible preference between one versus the other?
Yeah. Our preference is definitely to get take-or-pay arrangements. It's more stable, more predictable. Where we have the ability to secure those, that's what we prefer.
Who's more on the client side?
Oh, on the client side?
Yeah. What are they asking for?
Yeah. I think it depends on the client. I think we have a number of utility-type customers who have mandates to have firm transportation secured by their utility commissions and regulatory bodies. In those cases, I think they just want to have that security. They just want to have that commitment, and they're a little less concerned about the form of that commitment. In those cases, it's easy for us to obtain a certain length of take-or-pay contract. A number of our contracts, it's not optional. They have to take take-or-pay if they're going to have a contract with us. Otherwise, their services could be interruptible, and that's not desirable by a number of our customers. Really where you see the volume-based and fee-based business is less on our long-haul pipelines moving across multiple states.
It's more on the gathering and processing businesses that are involved with individual basins, that are gathering supply from individual basins. A number of our terminal businesses where that's the norm for that business, where we're storing liquids products, refined products, gasoline, diesel, and we're doing it for a fee. Some of those are more done on a volume-based business versus take-or-pay. Although we do have a significant number of our liquids terminals that are on monthly warehouse charges, which are effectively take-or-pay contracts.
We have time for two quick ones.
Sure. Does anyone else have any? Okay, go ahead.
In terms of the balance sheet, four and a half times EBITDA, why do you think it's the right number? If you were a private company, can you run the business like that with seven, eight times EBITDA? If you could comment on the build versus buy. Is it cheaper to buy a company or to build? How do you think about that in M&A? Thanks.
Yeah. If we were a private company, we would still be subject to the rating agencies weighing in on what the appropriate leverage is, and that would influence the cost. If we went much higher than our current level, we'd be subject to a downgrade. That's important to us because of the cost and the accessibility to the debt markets. While, as I mentioned earlier, we don't look to the debt markets to fund a number of our growth projects and our uses of cash. We do have maturities coming up on a regular basis. To refinance those, we want to make sure we have ready access and affordable access to the debt markets. That's one of the criteria that we put into the four and a half.
Our stable cash flow and predictable level of EBITDA is another important trait that we look at to determine what level of debt is right for us to be able to support. We're very comfortable at the 4.5x on that front. To go any lower than this, we don't see a significant benefit on our cost of capital or our overall cost of debt. We think this is a good spot for us. It's allowed us to achieve that mid-BBB level, which is an attractive rating for us. This gives us more cushion and more flexibility and better access, we think, more predictable access to the debt markets.
On the build versus buy, I think at these return levels, with these risk profiles of these projects that we have here, these are going to be economically superior to a buy, unless that buy comes at an unusually low price. I think that the trade-off is, if you're looking to achieve an established portfolio with immediate cash flows, sometimes buying is the only way to do it. Otherwise, build is usually the preferred method to go. Okay. All right. Thank you all very much. Really appreciate it.
Thank you very much.