愛依斯電力 (AES) 2017 Q4 法說會逐字稿

Good morning, and welcome to The AES Corporation Fourth Quarter and Full Year 2017 Financial Review Conference Call. (Operator Instructions) Please note, if you are listening to the webcast, please mute your computer speakers before asking questions. Please note, this event is being recorded. I would now like to turn the conference over to Ahmed Pasha, Vice President of Investor Relations. Please go ahead.

Thank you, Kate. Good morning, everyone, and welcome to our fourth quarter and full year financial review call. Our press release, presentation and related financial information are available on our website at aes.com.

Today, we will be making forward-looking statements during the call. There are many factors that may cause future results to differ materially from these statements. Please refer to our SEC filings for a discussion of these factors.

Joining me this morning are Andrés Gluski, our President and Chief Executive Officer; Tom O'Flynn, our Chief Financial Officer; and other senior members of our management team.

With that, I will now turn the call over to Andrés. Andrés?

Thank you, Ahmed. Good morning, everyone. Thank you for joining our fourth quarter and full year 2017 financial review call. During 2017, we delivered on all of our financial metrics. Adjusted EPS was $1.08 toward the upper end of our guidance range. Cash flow also came in at the upper end of our ranges.

Based on our strong performance in 2017, and our confidence in our outlook, we are reaffirming our 8% to 10% average annual growth rate through 2020. Further, we continue to transform and simplify the company. To that end, we are maximizing our efficiency with a new organizational structure, which will yield an additional $100 million in annual cost savings by 2019. We're reducing our financial risk by prepaying $1 billion in current debt. We're leveraging our platforms by adding 4.4 gigawatts of new capacity that is currently under construction.

Through a balanced approach, we've been reshaping our portfolio, while reducing our carbon exposure. First, by acquiring 2.3 gigawatts of renewables and launching the Fluence energy storage joint venture with Siemens. Second, we announced that we are selling or retiring 4.3 gigawatts of merchant, coal-fired generation.

Through the successful execution of our strategy, we're lowering the risk of our portfolio, particularly the volatility part of our earnings and cash flow. At the same time, we are well positioned to deliver 8% to 10% average annual growth in adjusted EPS and Parent Free Cash Flow through 2020, achieve investment grade credit metrics in 2019 and reduce our carbon intensity by 25% from 2016 to 2020. I will now discuss these themes in more detail, beginning with maximizing our efficiency on Slide 4.

We implemented a new $100 million cost savings plan as a result of our recently announced reorganization. This year, we are reducing our global workforce by 1,000 or 12%. These additional savings will strengthen our ability to deliver on our long-term financial commitments. Next, I'll provide an update on some of our construction projects.

In total, we have 4.4 gigawatts currently under construction. Most of which are expansions of our existing plants and businesses, beginning with Alto Maipo on Slide 5.

As you may recall, this 531-megawatt hydro project has been experiencing significant construction delay in cost overruns. However, since our third quarter call in November, we have reached a significant milestone towards resolving outstanding issues. Specifically, Alto Maipo negotiated a fixed price lump sum EPC contract with Strabag, the project's main contractor for the entire project. The new EPC contract, which is pending approval from the project lenders, transfers all of the geological risk to the contractor and includes material capital commitments from Strabag. The restructuring will require concessions from the project lenders and meaningful equity contributions from AES Gener, which are tied to construction milestones. We expect to receive approval from the lenders in the second quarter.

Although we are very disappointed with the extended delays and increased cost to build Alto Maipo, the new contract provides much greater certainty on both the schedule and the total cost to complete the remaining 38% of the project. Once completed, Alto Maipo will diversify AES Gener's generation mix and provide a zero-emission source of power and capacity in Chile's load center for many decades.

Turning now to the rest of our construction program beginning on Slide 6. Our 671-megawatt Eagle Valley CCGT in Indiana achieved full load earlier this month. This plant is now in the commissioning phase and is expected to be completed in the first half of the year.

Now turning to our 1.3-gigawatt Southland CCGT project on Slide 7, which is a new construction on our existing gas generation sites in Southern California. Construction is proceeding as planned and the project is on track to be operational by the first half of 2020. Shortly, we will also begin construction on this site on our long-term contracted 100 megawatt, 4 hour duration, lithium-ion, energy storage facility. This project will be the world's largest lithium-ion energy storage facility.

Turning to Slide 8 and our LNG businesses. In Panama, this month, we started commissioning at our 380-megawatt Colón CCGT. We expect to achieve first fire in March and COD early in the second half of this year. As you may remember, we're also building an LNG regasification and storage facility on the same site and expect to reach COD on time in 2019.

In the Dominican Republic, we are in advanced discussions to secure new clients for the excess capacity at our LNG storage facility and to build a pipeline to connect the LNG terminal to the eastern side of the island. The pipeline will allow us to sell our excess capacity at the existing plants -- as existing plants convert from heavy fuel oil and diesel to natural gas. We expect to earn attractive returns, given the limited amount of investment necessary and that the project will require no cash from corp. Our remaining construction projects are proceeding as planned, including our 1.3-gigawatt thermal plant, OPGC 2 in India. These projects will be key contributors to our earnings and cash flow growth through 2020.

Turning to Slide 9. We have been reshaping our portfolio to deliver attractive returns to our shareholders while reducing our carbon exposure. Our focus is on renewable projects, with long-term U.S. dollar-denominated contracts.

On a portfolio basis, these investments are expected to produce low- to mid-teen IRs, assuming a conservative terminal value. In general, we expect to receive at least 85% of the cash flow during the life of the PPA. These compelling returns are driven by several factors, including: About half of our investments are in markets with lower renewable penetration and faster growth rates in U.S.; using our business platforms and global scale to lower costs, such as PV panel and wind turbine purchases; utilizing local debt capacity in the businesses to fund the investments; and bringing in partners to reduce our equity commitments while providing management and development fees.

Turning to Slide 10. In the last year, we acquired 2.3 gigawatts of renewable capacity with long-term contracts in 3 markets: First, in the U.S. We closed on the acquisition of sPower together with the Alberta pension fund, AIMCo. sPower was a key driver in our 2017 growth and is continuing to execute on its more than 10-gigawatt development pipeline in the U.S. In fact, this year, sPower signed long-term PPAs for 582 megawatts of solar and wind capacity with investment grade customers; second, in Brazil. AES Tietê acquired 686 megawatts of long-term contracted wind and solar generation. The equity required for these expansions was funded by using the debt capacity available at Tietê; and third, in Mexico, where we have a 2.5-gigawatt development pipeline of renewables and natural gas infrastructure. We acquired the 306-megawatt Mesa La Paz wind development project. Mesa La Paz has a 25-year, U.S. dollar-denominated PPA with investment grade private sector offtaker. The project site has sufficient additional land to accommodate up to 200 megawatts of solar, which could be an attractive upside in the future. We expect to reach financial close in March and begin construction shortly thereafter.

During 2017, we also made good progress on our initiative to offer new innovative energy solutions. As a result, in Hawaii, we're delivering 2 solar, plus energy storage facilities for a total of 47 megawatts of solar and 34 megawatts of 5-hour duration storage on the island of Kauai. The first of these pioneering projects is under construction and will satisfy energy demand during peak hours in the evening as well as the rest of the day.

We also closed on Fluence, our joint venture with Fluence -- with Siemens. Fluence will deliver energy storage solutions and services to a broad group of customers, from commercial and industrial companies to utility and power developers around the globe. In fact, the team is currently pursuing more than 1 gigawatt sales opportunities in 15 countries. The goal is for Fluence to consolidate its position as market leader in this high-growth market. Lithium-ion-based energy storage is expected to grow tenfold in 5 years, reaching at least 28 gigawatts of global installed capacity by 2022.

In summary, as you can see on Slide 11, we will be adding 8.3 gigawatts of new capacity by 2020. This represents 25% of our current, installed capacity and includes 7 gigawatts of projects either under construction or recently acquired. The remaining 1.3 gigawatts reflect projects in advanced stage development, half of which are under signed contracts.

As a result of these additions, our average remaining contract term will increase from 6 years currently to 10 years by 2020. We have sufficient, internally generated cash to fund our equity contributions for all the projects I just discussed.

We're taking a balanced approach to decarbonizing our portfolio, recognizing that coal will continue to play a role. In 2017, we announce the exit of 4.3 gigawatts of merchant coal fire generation, representing 30% of our coal-fired capacity. Through these actions, we are significantly reshaping our portfolio to achieve our financial and strategic objectives.

As you can see on Slide 12, by the end of 2020, we expect our coal-fired capacity to decline from 41% to 29%, while renewables and gas will increase from 55% to 68%. Further, as you may have seen in our press release this morning and on Slide 13, I'm pleased to announce that based on these steps we've taken to date, we are on track to reduce our carbon intensity by 25% from 2016 to 2020 and we will be aiming for a reduction of 50% by 2030.

With that, I'll turn the call over to Tom to discuss our financial results, capital allocation, guidance and expectations in more details.

Thanks, Andrés. Good morning. Today, I'll review our 2017 results and capital allocation. I will also discuss recent business developments and conclude by addressing our guidance for this year and expectations through 2020.

As Andrés mentioned, we finished 2017 on a strong note, achieving the upper end of our guidance range on all metrics in setting a solid foundation for growth through 2020.

Adjusted EPS was $1.08. In the last 2 months of the year, we benefited from stronger margins at some of our businesses, a lower impact from hurricanes and a lower overall tax rate. As shown on Slide 15, most of our growth in 2017 was driven by higher margins, particularly in MCAC, contributions from new solar projects in the U.S. and the absence of a one-time reserve taken in 2016 in MCAC.

Now to Slide 16, our adjusted PTC and consolidated free cash flow. We earned a little over $1 billion in adjusted PTC during the year. This was an increase of 170 -- $167 million primarily due to the same drivers as adjusted EPS. We generated $1.9 billion of consolidated free cash flow, a decrease of $323 million from 2016, primarily due to large receivables collections in Eurasia and Brazil in '16.

Now I'll cover our SBUs in more detail over the next 5 slides, beginning on Slide 17. In the U.S., margins were flat. Adjusted PTC increased primarily due to equity earnings from new solar projects at sPower and our distributed energy business. Lower consolidated free cash flow also reflects higher working capital requirements at DPL.

Regarding sPower, we're very pleased with the business's performance since the acquisition. In November, sPower closed a $420 million, 19-year financing at 4.6%, enabling us to meaningfully increase our returns on the business. We also continue to receive inbound indications of interest at attractive valuations to partner on sPower's operating assets. Incorporating such a partner would further increase our overall returns and transition a greater percentage of our capital into sPower's robust development pipeline. This backlog continues to grow and is yielding excellent projects with double-digit returns, including the 580 megawatts of recently signed PPAs, Andrés mentioned.

In Andes, our results were relatively flat. Higher pricing in Argentina and a full year of operations at Cochrane in Chile, were largely offset by the impact of green taxes and planned major maintenance at AES Gener in Chile. Lower adjusted PTC also reflects higher interest expense in Argentina. Consolidated free cash increased largely due to lower working capital requirements at Gener.

In Brazil, margins were flat while adjusted PTC benefited from the settlement of a legal dispute at our CCGT, Uruguaiana, in the first quarter of 2017. The decrease in consolidated free cash flow is largely due to the high recovery in 2016 at Eletropaulo, a purchase power cost from prior droughts. Most importantly, as part of our strategic shift away from the distribution business in Brazil, in Q4, we reclassified Eletropaulo to discontinued operations. This reduces our volatility and eliminates the disproportionate exposure to Brazil in our consolidated financial statements given our 17% ownership interest.

For example, we've been consolidating over $3 billion of revenue with over $1 billion of unfunded pension liability with only $3 million of income in 2017. Mexico, Central America and the Caribbean, our results reflect improved margins, driven primarily by higher contracted sales in the Dominican Republic, following completion of the combined cycle last year as well as higher availability in Mexico. Adjusted PTC in '16 also reflects a reserve taken against certain reimbursements in MCAC in connection with a legal matter. Consolidated free cash flow also benefited from receivables collections in the fourth quarter in the Dominican Republic.

I'll also note that our plan in Puerto Rico is now being dispatched and delivering much needed energy to the grid. Payments from the offtake of PREPA have also resumed, and we've received $40 million since December.

Finally, in Eurasia, results reflect stable margins and the collection of a large overdue receivable in 2016 at Maritza in Bulgaria. Since the restructuring of Maritza's PPA in 2016, the offtakers have been paying on time. On the regulatory side, Maritza expects to have discussions, later this year, with the government of Bulgaria regarding the European Commission's review of the PPAs compliance with state aid rules. We'll keep you updated as discussions progress.

Now to Slide 22, an update on the impact of tax reform. As you know, we incurred a onetime, noncash charge of $1.08 in 2017 upon enactment of the new law, which was largely related to deemed repatriation of foreign earnings. This is a complex bill, and some issue still remain to be clarified. As we disclosed last month, in the near-term we expect a $0.05 to $0.08 annual impact, largely driven by 2 aspects: First, we expect meaningful limitation on interest deductions, which are now capped at roughly 30% of nonregulated U.S. EBITDA; second, under the new global intangible income rules, unrepatriated foreign earnings above a certain threshold can now be subject to U.S. tax. We've taken actions to offset these impacts and will continue to evaluate additional tax planning opportunities. In the longer-term, there's aspects of the tax reform that have been beneficial to AES. For example, the adoption of a territorial tax regime will provide more flexibility in structuring new investments and repatriating profits.

Now to Slide 23 and our improving credit profile. We ended '17 with a $4.7 billion of Parent debt and almost $2 billion reduction since 2011. As we announced in December, we use all the proceeds from the $1 billion Masinloc sale, to further reduce parent debt, which will bring our debt to about $3.8 billion. As a result, we now expect to achieve investment grade credit metrics in 2019, a year earlier than our prior expectations. We also have a high priority goal of attaining an investment grade ratings by 2020.

We believe this will help us to not only reduce our cost of debt and improve our financial flexibility, but also enhance our equity valuation. Now to 2017 Parent capital allocation on Slide 24. Sources on the left-hand side reflect $1.5 billion of total available discretionary cash, consistent with our prior expectations. This includes $637 million of Parent Free Cash Flow above the midpoint of our expected range.

Uses on the right-hand side of the slide are largely in line with our expectations. Investments and subsidiaries are slightly higher than our prior disclosure, largely due to additional investments in U.S. renewables.

Now turning to our guidance on Slide 25. Consistent with industry practice, these numbers exclude cost directly associated with major restructuring programs and the onetime, noncash charge of $1.08, resulting from the enactment of tax reform in 2017.

Today, we're initiating guidance for 2018 adjusted EPS of $1.15 to $1.25, and reaffirming our target of 8% to 10% average annual growth through 2020. Growth this year will be largely driven by contributions from new projects, cost savings and lower parent interest. To break this down by SBU, we expect growth in U.S. to be driven largely by positive regulatory actions at DPL as well as growth and renewables. Andes will benefit from continued market reforms in Argentina, higher contracting levels at Angamos in Chile and a higher generation in Colombia.

Growth in MCAC is expected to be driven largely by completed construction projects, including a full year of operations at our combined cycle in the Dominican Republic, as well as the partial year impact from the commencement of operations at the Colón CCGT in Panama.

Finally, we also expect to benefit from cost savings and lower Parent interest. This growth will be partially offset by business exits in the Philippines and Kazakhstan and the higher tax rate driven by U.S. tax reform.

Beginning this year, we'll no longer provide guidance on consolidated free cash flow, which does not accurately account for AES' ownership, interest in our underlying businesses. We believe that Parent Free Cash Flow to the most tangible measure are our ability to achieve our financial goals, including strengthening our balance sheet and delivering value to shareholders.

Turning to Slide 26. Parent Free Cash flow is expected to be relatively flat this year from $600 million to $675 million. This reflects lower expected distributions from Gener, to allow for incremental investments in Alto Maipo and ensure the maintenance of their investment grade credit ratings.

Consistent with prior expectations, we still expect 8% to 10% average annual growth through 2020 off the 2017 base. I'll now discuss our 2018 Parent capital allocation on Slide 27.

Beginning on the left side, sources reflect $1.9 billion of total available discretionary cash, including the $600 million to $675 million of Parent Free Cash Flow I just mentioned. Sources also assume $1.25 billion in asset sale proceeds, including the $1 billion sale of Masinloc in the Philippines and a $250 million placeholder for additional asset sale this year.

Regarding Masinloc, we recently received a key regulatory approval. The sale could close as early as the end of the first quarter. Now to uses on the right side of the slide. Including the 8.3% dividend increase we announced in December, we'll be returning $345 million to shareholders this year, as expected. We expect to use over $1 billion to reduce Parent debt, including revolver drawings.

Finally, we plan to invest at least $250 million in our subsidiaries primarily for projects under construction, leaving about $100 million of unallocated cash.

Now looking at our capital allocation from 2018 through 2020 beginning on Slide 28. We expect our portfolio to generate $4.2 billion in discretionary cash, roughly 60% of our market cap. This reflects Parent Free Cash Flow over the period and as well as our $2 billion asset sale target through 2020, half of which will be realized from Masinloc.

In terms of uses on Slide 29, more than half has been allocated to the current shareholder dividend in debt reduction. About $750 million is allocated to identified investments in our subsidiaries, including projects under construction in late stage development. The remaining $1.25 billion, which is largely weighted towards '19 and 2020, is available to create shareholder value through investment in compelling growth opportunities, modest deleveraging of about $100 million to $200 million per year and potential growth in our dividend.

With that, I'll now turn it back to Andres.

Thanks, Tom. Before we take questions, let me summarize the concrete steps we're taking to transform and simplify the company. Reducing our headcount by 12% this year for $100 million in sustainable cost savings, lowering our Parent debt by 20%, investing in profitable, renewable projects with long-term, U.S. dollar-denominated contracts, including the 2.3 gigawatts we acquired in 2017 and reducing our carbon exposure by exiting 4.3 gigawatts of merchant coal-fired generation.

Accordingly, as a result of our successful execution, we will deliver 8% to 10% average annual growth in adjusted EPS and Parent Free Cash Flow through 2020, achieve investment grade metrics in 2019 and reduce our carbon intensity by 25% from 2016 to 2020. Our overarching goal is to deliver sustainable and attractive risk-adjusted total returns to our shareholders.

Operator, we will now like to open up the lines for questions.

(Operator Instructions) The first question is from Ali Agha of SunTrust.

First question, Andrés, on Alto Maipo. In the past, when you have put that project as part of your construction pipeline, you assumed a 0 return on the investment you've already made there. Can you give us a sense of what's roughly the incremental investment you will need to make? And, sort of, the economic or financial case, internally, that you went through to decide, hey, we should go forward with this? As opposed to perhaps writing off the previous investment you've made there.

Surely. I can't comment right now on what the amount of the additional investment AES Gener would have to make at this point. That will happen when the financing is closed. But I would say that what is important is Gener has been strengthening its balance sheet, selling some assets. Also, as Tom mentioned, there will be somewhat less dividends this year as a result of investments to be made in Alto Maipo. Now in terms of what to do with the project going forward, of course, what matters is just the marginal cost and the marginal benefits. And that's what we have been looking at. I think that the -- from a strategic point of view, having Alto Maipo, plus Las Lajas and Alfalfal, it's all part of one big complex. You will have, at the end, 750 megawatts of capacity and power, right in the load center of Chile. So this, we think, will be a very attractive asset. And as you know, this will, say, balance Gener's risk because Gener has been heavily weighted towards coal plant. And as, again, Tom mentioned in his speech, we did have the impact of green taxes in 2017. So that's what we're looking at, Ali. We'll look at the entire, it's like the marginal returns on the investments we're making and the positioning of Gener as a company into the future.

Okay. Second question. As you mentioned, the Parent Free Cash Flow profile is relatively flattish in '18. How does that impact your dividend plans? And when you think about the 8% to 10% growth in dividends, would we expect that as an annual number or would that follow the Parent Free Cash Flow profile?

I think if you look at the growth of our dividend, I think we've had the fastest dividend growth of any company in our sector. We're -- and so I think, going forward, we'll continue to analyze what's the best use of our cash. Certainly, we don't think we're getting a lot of credit for our growth in the dividend, and the fact that we're on the path to become investment-grade. So we will look at what we think is the best use of our cash going forward. I think we've laid out -- our priorities is, one, to become investment-grade. And two, that we have the transformation of the company that's underway that will decrease risk and will also ensure our growth into the future.

And Andrés, just to clarify. Remind me again that 8% to 10% growth in EPS and Parent Free Cash flow, '18 through '20. Are you also committing to an 8% to 10% growth in the dividend commensurate with that?

At this stage, we haven't made any comment on an 8% to 10% growth in the dividend. We're committing to the adjusted EPS growth and the Parent Free Cash Flow growth.

I see. Last question, the $1 billion of asset sales that you are planning between the '18 through '20 period. Is the impact of that already factored into that 8% to 10% EPS growth that you've laid out for us?

Absolutely, absolutely. And so I mean, as I said, we've been talking about a balanced approach. I think if you look at our sales, we've really gotten good value for our shareholders on those sales. As I said, we also think that coal will continue to play a role. But what will we focus on? Certainly, we focused on coal merchant plants, going on. We've also focused on simplifying our portfolio. Now depending on the quality of the asset, whether it's accretive or dilutive, that will depend. But this is baked into our vision of the future and what we will deliver. So yes, we feel very comfortable about it hitting that $1 billion target. Some could be selling out and some could be selling down. And the other factor in our strategy, which we, perhaps, haven't highlighted in this speech, but we've attracted more than $4 billion of partnership capital. So we're really looking at maximizing our returns, and partnership capital in general has given us a lot of flexibility in the allocation of our resources, helped us manage risk and it's also helped us improve our returns.

Our next question comes from Angie Storozynski of Macquarie.

Okay, so 2 questions. How does the new solar panels impact your growth plans for sPower? And then, secondly, if you could elaborate a little bit about those negotiations concerning Maritza and some state aid that you mentioned.

Okay. Let's talk a little bit about the solar panels. The impact of the tariff, the 30% tariff, is baked into our forecast and, it hasn't had a material impact on the business. I don't know if, Tom, perhaps, you' like to comment on it?

No, it's fine, Angie. I think as Andrés said, the expectation of a tariff has been obviously out there for some time since early middle of last year. Some of the numbers being thrown around were actually higher than that. So it's certainly within expectations of what was going to be passed. I think that was taking into account by our teams at sPower and our distributed energy business. So in general, we've been moving forward. Yes, sure, on the margin, there were a small number of opportunities that kind of fell off or at least got put on hold, but then continues to moving forward. Well because -- as is evidenced by the large signing of contracts that we identified just most recently.

Okay. Let me take the second question, which was regarding Maritza. Well, we just wanted to mention this. At this stage, it's very early to say what the outcome of this would be. It's basically known that DigComp of the European Commission. Those reviews of long-term PPAs and whether they contain what's called illegal state aid. We have a -- we feel we have a very strong contract. We will -- there's a lot of ways that this could be resolved. And at this stage, we'll keep you informed as it progresses. But again we feel that we have a very strong contract. As you know, a couple of years ago, they ran up a very significant IOU, more than EUR 400 million. On our calls, we said look, we expect to be paid because of the strength of our contract and quite frankly, because of the investment-grade of Bulgaria and to the fact the public sector had the means to which to pay. So those things will resolve. Stay tuned. We'll see where this turns out. But again, we think we're starting from a strong position.

Our next question comes from Julien Dumoulin-Smith of Bank of America Merrill Lynch.

Hey, just wanted to follow-up on the last question a little bit. You alluded to or I suppose you reaffirmed today the 8% to 10%. Can you just clarify to the extent to which that, that is inclusive of the asset sales? Basically, that's sort of an implicit guide up because I think before you talked about the 8% to 10% being exclusive of those needing to find, call it, cost cut and/or other sources to offset the dilution from asset sales. Am I thinking about that correctly?

Yes, Julien, it's Tom. I think, maybe you want us to stay on the path. But I think no, the 8% to 10% is inclusive of asset sales. Once again, there's a baskets of things we may sell out of or sell down, as Andrés said. And it does also assume use of capital for deleveraging and reinvestment, I'd say, our reinvestment is at rates less than what we've been investing at. So we feel quite good about those.

Got it. Said differently again. You basically found the cost savings at this point to fully offset the full slate of 2018 through 2020 asset sales. And again, just to make sure that this is clear, that it's only $1 billion of asset sales that's reflected in that 8% to 10%? Or what magnitude through the full 3-year period?

No, it's only an additional $1 billion. So it's Masinloc plus $1 billion.

Our next question comes from Greg Gordon of Evercore ISI.

I have one other question. The language you used to describe the impact of tax reforms, you say, "Near-term impact of $0.08 to $0.10." What is -- does that imply that the impact, all things equal, and obviously, this is before what you're doing to offset it, changes over time? And if so, can you give us some insight into whether it gets better or worse over time and, sort of, a base case before offsets and then sort of other than the cost cutting, which you've announced, what are the things that you're doing to offset that sort of base case impact of tax reform?

So let me take sort of a high level and then I'll pass it to Tom. What we talked about is $0.05 to $0.08 initially. There are 2 sides to this: One is the, sort of, the limitations on interest, expense, deductibility related to EBITDA in the U.S; and the second is the global, intangible. Low tax income. Now there remains a lot to be clarified on this law. So this is -- we're taking a conservative approach to it. Why does this change over time? Well, it changes because of the -- your assets base changes and also, it has to do with your level of indebtedness changes over time. So what happens, over time, it gets better as the effects of a lower tax rate kick in. So that's number one. Number two, as Tom said in his speech, we really sort of cleaned the slate by basically using our NOLs to pay for the tax expense of deemed repatriation of foreign earnings. So really as a result of it, we'll have a much more transparent tax position as time goes by. Now we do need a further clarity, quite frankly, to make sure that any actions we take to optimize our capital efficiency are the right ones in the long-term. So with that, I'd turn it over to Tom.

Yes, Greg, I think that we say near-term, it's a 2- to 3-year period. We look beyond that, we're seeing lower impacts based on what we see in that -- as we see it. It gets -- there's a lot of moving parts. But in general, as we work through our NOL position, and as part of the charge we took at the end of the year and the deemed repatriation of foreign earnings, we used about $1.9 billion of our NOL. So as our NOL decreases when we move towards a taxable position over the next 2 to 3 years, the overall impact of tax reform from an earnings standpoint could actually be less. But that's what we say near-term $0.05 to $0.08, call it 2 or 3 years. And then as we look at it, we would see that number potentially going down. Obviously, we're trying to make those numbers lower and having the ramp-down effect accelerate if possible, but those are things that are still a work in progress.

Okay. And all that being said, that's fully baked into the growth rate expectations that you're aspiring to earnings and cash flow?

Yes. Yes, we have the $0.05 to $0.08 impact baked in through 2020.

The question is from Gregg Orrill of UBS.

Maybe you addressed this a bit earlier. But in terms of the -- your stance to keep AES Gener investment grade. What do you think is required there and what levers would you pull?

Well, we wouldn't -- basically been pulling the levers we talked about in the past. We said Gener has a lot of levers, realized that Gener had a very good year in 2017. So we've had some asset sales, sales from noncore assets. We have sold a peaking plant. And we're also in the process of selling some other noncore assets, which we think, quite frankly, will add very good multiples. And these will be used to pay down debt at Gener, to shore it up. So while the Gener price has suffered greatly the last, say, 2 years. At the same time, Gener in terms of, if you look at its earnings this year or its cash flow is really at record levels, given the fact that we cut the ribbon on time and on budget on the other projects. So Gener, again, we will keep it investment-grade. There are a lot of levers to pull. And as we said, we weren't putting in more money from AES into Gener.

I'll just say specifically, the dividend we got last year from Gener was around $160 million, $170 million. That was the AES share. In all the numbers, we're expecting that to be a lower number. I don't want to get specific and obviously, Gener is a public company. I'll just leave it that we are being more conservative with expectations for '18 until Alto Maipo gets tied up and moving forward.

The next question is from Steve Fleishman of Wolfe Research.

The question on the cost cuts and as you said, strengthening the 8% to 10%. Could you maybe just give a little more color on what you mean by strengthening? Is that -- are you kind of seeing yourself higher in the range or you have more cushion in the event that something doesn't go right? How should I think about that?

Yes. Well, I think when we talk about strengthening, it's quite frankly decreasing the band, increasing the certainty that you're going to hit your numbers. We had a lot of these things, I would say in the works, but it certainly feels good to have executed on them. So just to be clear on the cost savings program that we just announced. I mean, that's mostly executed. And we will have those numbers in hand in the first half of this year. Of course, we have the onetime severance costs as well included in our numbers. So most of this -- is that we feel that as we execute, we reduce the -- or increase, let's say, the certainty around our numbers and makes them more robust to any sort of unforeseen possibility.

And if you -- as you stand today, what do you see is the biggest risk to achieving the growth target, if any?

Well, I would say, look, we have, I think, listed them. I think it's very important to close on Alto Maipo. So we have -- even though it does mean Gener has to make material contributions, this is a very dramatic shift in the risk of this project. Because we're going from one, where essentially, when it came to geological risk, it was cost-plus to one where we have certainty on the cost. And second, where the -- we really have 1 contractor rather than 2, and that, that contractor will have significant skin in the game in every incentive to efficient -- finish at the lowest cost and as early as possible. So I think that, that is a very important. I think as I mentioned, regarding Bulgaria, this is something that we have to see the resolution of it. We think we're in a very strong position like we thought we were when we ran up the $400 million of -- or EUR 400 million, excuse me, of IOUs. But stay tuned because it's too early to say where that would turn out. I think that we have opportunities for the upside in a number of our markets. I'm very encouraged by the returns we're getting on our renewables and gas projects, especially outside the U.S. We have a excellent pipeline of projects in Mexico, dollar-denominated with private sector offtakers. Very excited about the solar plus energy storage possibilities. Very excited about the sPower pipeline. And regarding fluids, I think we're seeing that, that market is starting to turn. This is a classic new technology, which will go through an S curve. And we have to see when that turns up. Now speaking about Fluence, I think that's more of a play where it's not sort of incremental earnings per year, it's just really creating a lot of value, 3 to 5 years from now, probably, like we did with the Brazilian telecom.

Okay. And then last question just on the -- Tom, I think you mentioned something about a potential partner for sPower. Could you give a little more color on that? And just are -- is this mainly for kind of somebody who would buy a stake in current operating projects?

Yes, Steve, we've been approached by some parties about buying stakes and operating projects. It would be something that we would do obviously with us and with AIMCo jointly. I think, it just reinforces the value of their operating portfolio, we're obviously quite focused in the growth portfolio. But it reinforces the value of their operating portfolio and it's potentially attractive to some co-investors. So we'll see.

And that would effectively be part of your asset sales?

Yes. I mean, it's part of a bucket of opportunities -- basket of opportunities, yes. Instead of opportunities, we've -- $2 billion, we've got $1 billion as Masinloc of the $1 billion set. The set of opportunities is greater than $1 billion.

The next question is from Charles Fishman of Morningstar Research.

Just specific, let's say, I guess I'm looking at Slide 52, where you break down the adjusted PTC for '18 by SBU. And Andes certainly had some strong growth, '18 versus '17. And I realize you don't -- you're not giving -- when you talk about 8% to 10%, you're not breaking it down by SBU. But certainly, there are some very significant growth opportunities that you've outlined in the U.S. as well as MCAC. I sort of thought that maybe with the problems at Alto Maipo, it would take the wind out of the sails of the growth in Andes. And yet you had good growth in '18. Am I being too hard on -- are the other opportunities in that SBU just greatly outweighing the problems at Alto Maipo? Or is Alto Maipo not as big a hit as I thought then I realized you haven't disclosed that. Where am I going wrong on my outlook for Andes?

Sure. Let me clarify. First, remember that again, Gener has had a record year last year. The Alto Maipo issues, are sort of perspective. It's not current. The drop in the price of Gener stock is portably sort of, I think, a decrease in prospective prices, the market is thinking, what are the future prices? Now realize Gener is fully contracted through 2023 and 50% thereafter. Realize also that Gener is taking a very large market share of commercial and industrial contracts. So Gener is a strong company, an investment-grade company. Now talking about Andes itself, we had a very good year in Argentina. We had 3 gigawatts of excellent assets in Argentina, we've always said. We've have a relatively low debt in Argentina. So to the extent that the wholesale market has liberalized in Argentina and dollarized, this has been a positive for us. Also, realize that AES Gener is about 30% Columbia, 30% hydro in Chivor. So Andes is SBU. It's much more than just Alto Maipo. And realize that you're talking about, in total, almost 7 gigawatts of capacity that you have in Andes and Alto Maipo being 500. So the important thing, I think, is to resolve Alto Maipo to decrease the uncertainty and I would expect to have a positive reaction in terms of Gener stock and hopefully, AES stock as well.

So really, Andes will contribute -- again, realizing you're not breaking the 8% to 10% down by SBU. But do you foresee Andes contributing to that 8% to 10% just as -- along with the U.S. and MCAC, correct?

Yes, I'd say, certainly, we don't see it as a drag. We have interesting renewable opportunities in the region. And that even Gener has solar projects now under construction. So there's a lot of opportunities in the region. And what we're looking forward to is resolving the issues of Alto Maipo and have certainty in getting past that and really focus on the other projects in the future.

Okay. Well, sounds like your team has done an excellent job of being close to resolving a difficult situation in Alto Maipo. So that's certainly a positive.

You next question is from Lasan Johong of Auvila.

Andrés, I'm a little confused. In the press release, it says that AES is classifying Gener -- not Gener, Eletropaulo as a discontinued operation and then Tom said that it's going to be deconsolidated. My understanding was that it was going to be deconsolidated, not sold. So is there a change in status of Eletropaulo?

It's both deconsolidated and classified under discontinued operations. So that would be, as I said, it would be part of the -- it's true on both counts. In other words, our revenues and operating costs and everything don't flow through our financials, which may serve financials much easier to understand, especially given the depth at the end of unfunded pensions and those kinds of things. But also, as classified as discontinued operations that would say that we're going to continue on our strategic shift with respect to Eletropaulo and assess our ownership interest.

So at some point, it would be up for sale?

Yes, I don't want to get too specific. It's a public company. But that's being, kind of -- by being a discontinued operations, you can read from there.

Make our own assessment. Yes, okay. Andrés, it sounds like AES is moving much more towards a carbon-free portfolio in the U.S. And so it's interesting the development of sPower going forward. Would it be advisable at some point for AES to sell IPL and DPL and use that capital to bolster both sPower's development program and maybe do further renewable acquisitions in the U.S.?

Yes, I think -- thanks for the question. To me -- put a little bit on sort of carbon derisking in perspective. So we're taking a balanced approach. We can see that coal could have a role to play in some markets well into the future. If you take our U.S. operations, for example, IPL. IPL will go from 79% coal to about 44% coal when we cut the ribbon in Eagle Valley. So what we're really talking about is we see this is a long-term sort of derisking. We also will be doing things like -- we have found ways to run our coal plants at lower [maintenance]. We're talking about taking them down from 50%, 40%, down to around 20%. So this, combined with at certain hours of the day, very cheap renewables, will allow us to run our -- decrease our carbon footprint. But the same time, take advantage of our coal plants using basically like large batteries. So when you talk about something like IPL, we think it's -- or DPL, these are part of the strategy and that they really complement what we are in terms of the various sources of financing we have and the various growth opportunities. So right now, we would consider those utilities as core to our business proposition.

Understood. Last question to Tom. The interest deduction -- restriction on the U.S. portion. I'm doing a back of the envelope calculation. It looks like about $20 million to $30 million would not qualify out of about $265 million. Is that about right? I mean, are we in the same familiar ballpark?

Yes. Just on a stand-alone basis, hard to look at the tax -- any tax piece of the puzzle in isolation. But our U.S. income that would be available to shelter our interest would be less than that. Keep in mind that the U.S. income that's in the utilities and some of the other investments is not income that's available to offset interest. But it's a longer story than that. I think...

All right, it's just unregulated stuff.

Yes. But it's, as always, it's not that quite simple. I would say our interest is coming down meaningfully. Our interest is -- last year, it was about -- it was well over $200 million. I think as we see it after paying down the $1 billion of debt and also look at some other opportunities, we'll be under $200 million on a run rate by midyear. So next year, we expect interest to be maybe $180 million, something in that ballpark. So we're -- by reducing debt, we're meaningfully limiting the issue.

Okay. So next year, how much of the interest expense would not qualify for the deduction?

Yes, I'd rather that -- I think there's a lot of moving parts that's why we just try to bake it all together and said $0.05 to $0.08. And once again, it's hard to look at. Even some of the foreign -- some of the issue we talked about excess taxes on foreign, that can be an offset in part to the interest. So there's a lot of different equations. But just to boil it down, we think $0.05 to $0.08 for the next 2 to 3 years.

This concludes our question-and-answer session. I would like to turn the conference back over to Ahmed Pasha for any closing remarks. The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.