Feb 26, 2015
Executives
Ahmed Pasha - Vice President of Investor Relations Andrés Ricardo Gluski Weilert - Chief Executive Officer, President, Director and Chairman of Strategy & Investment Committee Thomas M. O'Flynn - Chief Financial Officer and Executive Vice President
Analysts
Ali Agha - SunTrust Robinson Humphrey, Inc., Research Division Stephen Byrd - Morgan Stanley, Research Division Julien Dumoulin-Smith - UBS Investment Bank, Research Division Maura A. Shaughnessy - Massachusetts Financial Services Company Greg Gordon - Evercore ISI, Research Division Insoo Kim - RBC Capital Markets, LLC, Research Division
Operator
Good morning. My name is Sean, I will be your conference operator today.
At this time, I would like to welcome everyone to the AES Corporation's Fourth Quarter and Full Year 2014 Financial Review Conference Call. [Operator Instructions] Thank you.
Vice President of Investor Relations, Mr. Ahmed Pasha, you may begin your conference.
Ahmed Pasha
Thank you, Sean. Good morning, and welcome to our Fourth Quarter and Full Year 2014 Earnings Call.
Our earnings 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 Andres 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.
Andres. Andres?
Andrés Ricardo Gluski Weilert
Good morning, everyone, and thank you for joining our fourth quarter and full year 2014 earnings call. Today I will, first, review our 2014 results; second, discuss the current macroeconomic environment and how it will affect our 2015 guidance; third, provide an update on the execution of our strategy to date; fourth, share my thoughts on capital allocation; and fifth, discuss our priorities for 2015.
Then Tom will discuss our 2014 results and 2015 guidance and longer-term expectations in detail. Turning to Slide 4.
During 2014, we made significant progress on our strategy and continued to position our company for the future. During this year, we brought in financial partners to invest $1.9 billion in our subsidiaries; announced or closed 10 transactions for $1.8 billion in equity proceeds from asset sales; broke new ground on 6 new platform expansion projects, totaling 2,200 megawatts and won long-term contracts to build 1,400 megawatts of capacity in California.
We allocated $600 million to reduce parent debt and improve our credit profile, returned $450 million to shareholders through dividends and buybacks and announced a doubling of our dividend, with an intended growth rate of 10%. Turning to Slide 5.
Unfortunately, our financial results for the year were affected by $0.10 of adverse hydrology in Panama and Brazil. Still, we earned an adjusted EPS of $1.30, which was at the lower end of our original guidance range of $1.30 to $1.40, and slightly better than our expectations at the end of our third quarter call in November.
We are disappointed with our proportional free cash flow of $891 million. Although proportional free cash flow came in at the low end of our revised guidance, it is 20% lower than our original guidance, primarily driven by the higher working capital requirements in Brazil and Chile as well as increased receivables in Bulgaria.
All of which, we expect to reverse in 2015. Now turning to Slide 6.
I would like to outline several factors affecting our 2015 outlook. Importantly, we have taken a number of steps to mitigate their impacts.
Since our last call, we've seen international macroeconomic factors move against us. Currency and commodity forwards have declined significantly.
Although we are largely contracted and the majority of our earnings are U.S. dollar-linked, this downward shift in forward curves has affected some of our businesses and, consequently, our earnings expectations for 2015 and beyond.
We are also seeing continuing poor hydro conditions in Brazil, especially in the state of São Paulo, where our hydros are located, rather than a return to normal hydrology, as we had previously assumed. We've been taking actions to lower our sensitivity to hydrology by adopting more optimal hedging strategies in Panama and Brazil.
In Panama, we are bringing in a 72-megawatt oil-fired barge. In Brazil, we are working on creative solutions to supply additional energy, such as pursuing opportunities to restart long-term operations at our 640-megawatt Uruguaiana CCGT plant, which has not operated continually for many years due to lack of fuel.
We're also exploring options to export energy from Chile and Argentina to the Brazilian grid. In parallel, we are proactively hedging our FX exposure in Brazil, Colombia and Europe, where since our last call, we have entered into additional hedges to shield 40% of our exposed earnings from further volatility.
In our 2015 guidance, we're also assuming that ongoing negotiations in Bulgaria will have some earnings impact on our Maritza business. The Energy Minister of Bulgaria recently issued a communiqué in which he committed to paying Maritza's outstanding receivables of approximately $260 million and announced ongoing discussions to reduce the contract price.
Additionally, the government is taking steps to improve the financial position of our offtaker, NEK, by reducing the volume of expensive energy that NEK is purchasing from other market participants and compensating NEK through environmental taxes. The Bulgarian government has targeted closing this negotiation and addressing NEK's financial situation in the first half of this year.
Although negotiations have not been finalized, we're incorporating modest earnings dilution in our future expectations, which we believe is sufficient to accommodate the outcome of the discussions currently under way. The combined earnings impact of the macroeconomic factors and hydrology in Brazil as well as the potential outcome of the PPA negotiations at Maritza is approximately $0.18.
Through proactive steps, including additional hedges, revenue improvement and cost savings, we expect to offset $0.13 of the total and, therefore, are reducing our adjusted EPS guidance by $0.05 to a revised range of $1.25 to $1.35. Notwithstanding the impact on our earnings from these factors, we are also reaffirming our 2015 proportional free cash flow range of $1,000,000,000 to $1,035,000,000 (sic) [$1,350,000,000], which is 30% higher than our 2014 results.
This growth is largely driven by the recovery of working capital in Brazil and Chile as well as a reduction in accounts receivable in Bulgaria and the contributions from new plants coming online this year, such as the 1,240-megawatt Mong Duong project in Vietnam and the 72-megawatt oil-fired barge in Panama. Turning to Slide 7.
I will discuss our continued progress on our strategic objectives, which we laid out in 2011, including reducing complexity, performance excellence, expanding access to capital and leveraging our platforms. First, reducing complexity.
Since 2011, we have reduced the number of countries where we operate, from 28 down to 18, and raised $3 billion in equity proceeds from asset sales. In 2014 alone, we closed 10 transactions, totaling $1.8 billion in equity proceeds to AES.
In addition to simplifying our portfolio, we recently exited riskier markets, such as Ukraine, Nigeria and Cameroon in a timely manner and at attractive valuations. As a result of our efforts, 80% of our 2014 earnings and proportional free cash flow was generated in 10 countries in the Americas.
Regarding performance excellence, we believe that we are now the low-cost manager of a large portfolio of international energy assets. As you can see on Slide 8, we have reduced our global G&A by about 1/3 or $200 million, achieving the goal we established in 2011, 1 year early.
Going forward, we are focusing on additional cost savings initiatives, including saving $100 million in O&M by 2018. Turning to Slide 9.
Through financial partnerships, we are expanding our access to capital and fine-tuning our portfolio's global macroeconomic exposures and commodity risks. In most cases, we earn management, development, a promote or upfront fees.
Partial sell-downs of our assets also served to highlight the value of the business in our portfolio. In total, we have raised $2.5 billion in proceeds to AES through financial partnerships.
In 2014, we brought in partners at 4 of our businesses, including CDPQ, a long-term institutional investor headquartered in Québec, Canada, which recently invested in IPALCO in Indiana. We look forward to working with CDPQ on additional partnering opportunities in the U.S.
and Latin America. Turning to leveraging our platforms on Slide 10.
We are exclusively focusing our growth on classroom expansions, including adjacencies such as energy storage and desalinization. Adjacencies are smaller investment opportunities that are replicable across our portfolio and have higher returns, with a much shorter construction period.
To that end, we recently made a $25 million investment to acquire Main Street Power, a developer of distributed solar. Although modest in size, this business will provide us with the know-how to implement distributed solar generation in some of our international markets, where power prices are higher and solar resource is greater.
We will focus on commercial and industrial customers. We currently have a total of 1,141 megawatts (sic) [7,141 megawatts] under construction, the most in AES's 34-year history.
These projects represent $9 billion in total capital expenditures, the vast majority of which is being funded by a combination of nonrecourse debt and partner equity. More importantly, our required equity for these projects is $1.5 billion, of which we've already funded 70%.
In terms of where these projects are located, as you can see on the right-hand side of the slide, 40% of the capacity under construction is in the U.S. The 1,400 [ph] megawatts of Southland contracts we recently won are not yet included in these numbers since we have not yet broken ground.
If they were, the U.S. would represent 50% of the new capacity.
We're earning very attractive risk-adjusted returns on these projects. For competitive reasons, we cannot provide details on a project-by-project basis.
However, we will provide some general guidelines. For international projects, we're seeing IRRs from the midteens to more than 20%, with project-specific returns varying, reflecting project and country risk premiums.
For U.S. projects, IRRs are in the low double digits.
Including all of our projects under construction, the average IRR is around 14% and the ROE is greater than 15%. We would expect to earn at least as much from new projects in our development pipeline.
Now turning to Slide 11. I would emphasize that as we have done in the past, we will complete all new projects with share buybacks and, furthermore, we will only invest in a new project if it meets the following criteria.
First, it enhances the value of an existing business, such as the MATS CapEx program at IPL in Indiana or the oil-fired barge in Panama. Second, it offers compelling risk-adjusted returns while minimizing AES Corp.
equity by using project-level cash or local leverage capacity. Examples include closing the cycle at our DPP plant in the Dominican Republic and energy storage.
And finally, for any large project, we would expect to bring in a partner to maximize our returns and allow us to fine tune our total exposure in the project. The bottom line is that our successful execution on the 4 pillars of our strategy that I just discussed have positioned us to deliver average annual cash flow growth of 10% to 15% over the next 4 years, as our construction projects come online.
We expect to grow our dividend 10% per year from today's level as cash flow increases. Given all that, we believe that our current share price does not reflect the progress we have made in our company and portfolio or the value from our largely funded construction program.
Therefore, we have taken advantage of low share prices by buying back $150 million of our shares since our third quarter call. And today, we have announced that our board has authorized an additional $400 million for share repurchases, the majority of which we expect to utilize in 2015.
Finally, as you can see on Slide 12, our overall capital allocation over the last 3 years has been very shareholder-focused. In fact, we have allocated 78% of our discretionary cash to parent debt prepayments and returning cash to shareholders.
Specifically, we've allocated $1.6 billion to decrease our parent debt by 20% and improve our financial flexibility. We have also reduced our share count by 10%, buying back 78 million shares at an average price of $12.69.
And with the recent doubling of our dividend, we're now paying $0.10 per share per quarter and we expect to grow the dividend 10% annually. We recognize that 2015 will be a challenging year due to negative macroeconomic factors in international markets and poor hydrology in Brazil.
Nonetheless, we will continue to execute on our strategy to create shareholder value by pulling the levers we have outlined today. With that, I will now turn the call over to Tom, who will provide greater detail on our 2014 results and 2015 forecast.
Thomas M. O'Flynn
Thanks, Andres, and good morning, everyone. Today, I'll review our fourth quarter and full year results, including: adjusted EPS; adjusted pretax contribution, or PTC, by strategic business unit, or SBU; and our cash flow.
Then I'll discuss our 2014 and '15 capital allocation plans, 2015 guidance and our longer-term expectations. As Andres has just discussed, we achieved a number of our objectives in 2014, despite the impacts from adverse hydrology and other factors.
From an EPS perspective, we were able to offset these issues through operational improvements, accelerated cost savings and our capital allocation decisions, including share repurchases and debt prepayments. Admittedly, our 2014 proportional free cash flow was disappointing due to higher working capital requirements and increased receivables.
Turning to Slide 14. Our fourth quarter adjusted EPS was $0.41 compared to $0.29 in the fourth quarter of 2013.
Overall, our businesses contributed $0.08, driven by higher earnings from our U.S., Brazil and Andes SBUs. We also benefited from a 3% reduction in share count, lower parent interest expense and improved corporate charges, which contributed $0.07.
These benefits were offset by a $0.03 impact from tax. The adjusted effective tax rate in the quarter was 25%, which was higher than the 18% rate we had last year.
Detailed drivers of our fourth quarter PTC performance by SBU are in the appendix. Moving to Slide 15, and our full year EPS of $1.30 per share.
Compared to 2013, we benefited from $0.07 of better operating performance at our U.S. SBU and our utilities in Brazil, plus a smaller impact from hydro; $0.03 from several other items, including recognition of higher interest income on receivables in Argentina in 2013; and $0.11 from capital allocation and lower corporate charges.
These benefits were offset by $0.06 from outages in the first 9 months and a $0.14 impact from a higher adjusted effective tax rate of 30% versus 21% in '13. We ended up at $1.30, which was at the low end of our original guidance range.
Turning to Slide 16. Our results by SBU were in line with the adjusted PTC modeling ranges we provided in November.
The main driver that helped us come in at the high end of our revised EPS range was a 30% tax rate, which was slightly below what we were expecting, and equates to a couple cents of earnings. Now to Slide 17.
Poor hydrology in Latin America continued to be a headwind in 2014, with a $0.10 impact to our adjusted EPS versus normal conditions. Although this was a meaningful impact, it was about 25% lower than what we experienced in 2013 due to favorable hydro conditions in Colombia and our proactive steps in Panama.
Now going to hydrology by market, beginning with Colombia. In 2014, Chivor had favorable inflow conditions relative to the rest of the country, which allowed us to sell excess power into the spot market at attractive prices, adding $0.03.
Currently, inflows at Chivor are 92% of average. And we're expecting a normal year in 2015.
In Panama, we had a $0.06 shortfall in '14, most of which was during the first 9 months, as inflows improved in the fourth quarter. Inflows are now close to 100% of long-term average and power prices are about $100 a megawatt hour, down from well over $200 per megawatt hour from much of last year.
Our downside exposure continues to decrease for 2 main reasons. First, the government compensation agreement we negotiated last year to offset this exposure; and secondly, lower oil prices helped to reduce our earnings volatility, as spot prices during dry conditions are set by oil-fired generation.
In 2014, we were purchasing in the spot market. If oil prices were at today's level last year, the hydro impact would've been about $0.03 rather than the $0.06 we actually experienced.
I'll also note, as Andres mentioned, we've brought an oil-fired barge to Panama, improving our bottom line. Finally in Brazil, we had a $0.07 impact from poor hydrology in 2014.
Tietê experienced a 10% reduction in generation, or GSF. The impact was magnified by the fact that the average spot price was just under BRL 700 per megawatt hour in 2014, significantly greater than the BRL 200 level of our offtake contract.
Currently, reservoir levels are about 23%. While we've seen some recent improvement, it's difficult to predict how much they'll recover by the end of the rainy season in April.
For now, we're factoring in about a $0.05 impact into our 2015 guidance due to an estimated 15% to 17% reduction in generation for the year. The impact of a shortfall will be less than last year as spot prices are now capped at BRL 388 per megawatt hour compared to the cap last year which was BRL 823.
We're not incorporating rationing into our 2015 guidance at this time. We'll provide an update on our next earnings call in May, when we'll have more clarity.
Although it's very premature, we estimate a rationing scenario could result in an additional $0.05 impact, above the $0.05 we already have incorporated into our guidance. Now to cash flow, beginning on Slide 18.
We generated $287 million of proportional free cash flow in the fourth quarter, bringing the full year amount to $891 million versus $1.27 billion in 2013. Although near the bottom end of our revised guidance, it's roughly $250 million less than our original guidance.
We're obviously disappointed with this result. This decline was largely driven by 3 unanticipated factors: first, an increase of a little more than $100 million in receivables at our Maritza Plant in Bulgaria; second, higher receivables at Sul, our utility in Brazil, due to higher energy purchases; and finally, a lag in VAT, or value-added taxes, related to our construction projects in Chile.
We expect improvement in these factors this year. Now to Slide 19, and our capital allocation in 2014.
Starting on the left, we generated roughly $1.9 billion of discretionary cash, including $1.2 billion from announced asset sales. We're pleased that despite our shortfall in proportional free cash flow, we generated $523 million in parent free cash flow, above the $500 million midpoint of our expectations.
Turning to the right-hand side of the slide, we allocated 76% of this cash to parent debt and returning cash to shareholders through buybacks and dividends. Meeting our parent free cash flow expectation is especially important as it is the foundation of our dividend and dividend growth.
As we've demonstrated through the recent doubling of our dividend, we continue to be comfortable that we'll generate about $525 million in parent free cash flow in '15 and growth of 10% to 15% annually. Turning to Slide 20, and our capital allocation plan for 2015.
With $460 million in announced asset sale proceeds, our total discretionary cash is expected to be about $1.5 billion in 2015. We'll invest about $300 million to fund equity commitments in projects currently under construction, the largest of which is at IPALCO, where we have a $1.4 billion investment program under way to increase the rate base by 70% from 2013 to 2017.
After taking into account our growth investments as well as expected debt prepayments of $200 million, we're projecting about $600 million of remaining cash available for allocation. We currently have a $400 million authorization for share repurchases, of which we expect to utilize the majority during 2015.
Before turning to our guidance, I'd like to take a moment to address our foreign currency and commodity exposures and our hedging practices, on Slide 21. We have long-term protections built into our business model to manage our exposure to FX and commodity price changes.
For example, 82% of our businesses are either utilities or generation businesses with an average contract term of 7 years. This insulates the majority of our earnings from commodity price volatility.
Having said that, some of our businesses do have commodity exposure as they operate in markets where power prices are oil- or natural-gas-driven. As shown in our appendix, every 10% move in oil is roughly a $0.025 EPS impact on an op [ph] basis, primarily from the Dominican Republic and, to a lesser extent, Panama, where we have a modest amount of excess power to sell into the spot market.
On the plus side, in Panama, as I just mentioned, lower oil prices reduce our exposure to poor hydrology. In terms of gas, our largest exposure is at our DPL merchant business, where our 3,000-megawatt coal-fired fleet is negatively impacted by declining gas prices.
However, although, gas has fallen by 20%, dark spreads held up at around the $12 range. The impact on our EPS has been less than our disclosed sensitivities, which assumed a constant market heat rate.
Regarding currencies, as you can see on the right-hand side, we're largely protected from foreign currency changes, given that more than 2/3 of our PTC is effectively U.S.-dollar denominated. The remaining 1/3 is exposed to a handful of foreign currencies, mainly the Brazilian real, the euro and the Colombian peso.
For the exposed portion, we mitigate the risk on a 12-month rolling basis through shorter-term FX hedging programs or currency-linked escalation clauses in our shorter-term contracts. Now I'll walk through our 2015 adjusted EPS guidance on Slide 22.
The updated forward curve on commodities and currencies used in sensitivities we provided in our last earnings deck, would suggest a $0.10 negative impact to 2015. However, as a result of additional FX hedges, fuel and power contracts, we have been able to trim the overall impact from currencies and commodities on our 2015 EPS by $0.04.
We're also incorporating a $0.05 impact from Brazil hydrology, as I discussed earlier. Finally, we've also factored in a modest dilution of $0.03 from other factors, including PPA negotiations at Maritza.
We're offsetting these impacts through the following actions: $0.04 from cost savings initiatives across our portfolio and revenue improvements, including a modest uplift from Mong Duong, where a new accounting standard better aligns earnings with cash on an ongoing basis; $0.02 from accelerated capital allocation; and $0.03 from various tax opportunities. Net-net, we've offset $0.13 of the $0.18 of headwinds, and therefore revised our guidance downward by $0.05 to a range of $1.25 to $1.35 per share.
Turning to our longer-term adjusted EPS expectations, on Slide 23. We continue to expect modest to flat growth in 2016, and 6% to 8% growth in 2017 and '18, more weighted towards 2018, consistent with our prior expectation but off a lower 2015 base.
Our projects under construction will drive our earnings growth, while many of the other actions we're implementing in 2015 will continue to benefit us in '16 and beyond, largely offsetting what appears at this time to be a negative impact from currency and commodity forward curves as well as the impact of potential modifications of the Maritza PPA. Shifting to cash flow guidance on Slide 24.
We are reaffirming our 2015 proportional free cash flow guidance of $1 billion to $1.35 billion. This implies an increase of roughly $275 million over 2014, which is driven by recovery of working capital and receivables along with cash flow from new businesses, including Mong Duong, which comes online this year.
In Brazil, we're encouraged that recent moves by the regulator will accelerate the collection of purchased power costs by the distribution companies. On Slide 25, we continue to expect 10% to 15% average annual growth in proportional free cash flow driven by contributions from projects currently under construction as well as the fact that maintenance CapEx is less than depreciation from new projects.
In 2014 and '15, depreciation averages about $350 million greater than maintenance and environmental CapEx. We expect this differential to widen to $600 million by 2017, which explains much of why our cash flow is growing at a greater rate than earnings.
We believe cash flow is a key value driver, with strong double-digit growth in both proportional and parent free cash flow over the longer term. We'll continue to show the value of our cash flow through disciplined capital allocation that maximizes risk-adjusted returns to our shareholders.
Now I'll turn it back to Andres.
Andrés Ricardo Gluski Weilert
Thanks, Tom. To summarize, we made significant progress on our strategy in 2014 and are encouraged with the resilience of our platform and the opportunities it provides to increase shareholder value.
For 2015, our priorities are: pulling all levers to mitigate the effects of poor hydrology in Brazil and lower FX and commodity prices; completing the 1,240-megawatt Mong Duong project in Vietnam, which will be a major contributor to our growth; resolving Maritza's outstanding receivables and renegotiating our PPA; executing on platform expansion opportunities and bringing in financial partners at the project level; and allocating our discretionary capital to maximize returns, including providing for growth in our dividend and returning cash to shareholders through share repurchases. We believe that AES is uniquely positioned to offer our shareholders a strong and growing cash flow and dividends.
I would like to open up the call for questions.
Operator
[Operator Instructions] Your first question comes from the line of Ali Agha from SunTrust.
Ali Agha - SunTrust Robinson Humphrey, Inc., Research Division
A couple of questions. One, I noticed that when you updated your 2015 guidance, you used the commodity and currency curves as of December 31, 2014.
As you know, since then, particularly on the currency side, we've seen further depreciation out there. And I'm just wondering, factoring all of that in as we stand here today, does that cause you to relook at the range of earnings that you put out maybe towards the lower end or has there been other offsets that you're still comfortable on the midpoint of this range to [indiscernible] given the currencies moves for the year?
Andrés Ricardo Gluski Weilert
Ali, we're comfortable with the range we're providing.
Ali Agha - SunTrust Robinson Humphrey, Inc., Research Division
Are there any other additional offsets that you've made since December 31 that aren't?
Andrés Ricardo Gluski Weilert
Again, I think that we have a number of initiatives that we're undertaking. And again, in the net-net, we're comfortable with the range we're providing.
Thomas M. O'Flynn
Yes, Ali, I'll just say that as we look at the currencies maybe off $0.01, maybe real is a bit up, but there are some other moving pieces. Maybe commodities have helped us a little bit, so net-net, we're in about the same place.
And we thought that the end of the year was a good baseline, and that's what we used for our budget. We just thought that was a good place to stick with, but we obviously provided you with sensitivities.
Ali Agha - SunTrust Robinson Humphrey, Inc., Research Division
Right. Second question, with regards to the longer-term earnings outlook, '16 and beyond, if I heard you right, I think you've said, "Hey, we're now going to start off from the lower base and build out from there."
So I would've thought some of the '15 hits are not dominant and currency and commodities can move at other places. So why is there a permanent sort of rollover effect in the future years that causes everything to move down, given the '15 move down?
Andrés Ricardo Gluski Weilert
Ali, I completely agree that currencies and commodities can move. Hydrology, we expect to be sort of onetime hits.
But currency and commodity, we're putting in the forward curves that we have to date. We don't have any better information.
But I agree, if they improve and then we would improve as well. But the way the modeling works is that you have to take the forward curves and the best estimate of tomorrow's price is today's prices, and so we've moved them down accordingly.
But I agree with you. I mean, one thing we're sure of is that they're going to move in the future.
Ali Agha - SunTrust Robinson Humphrey, Inc., Research Division
And then, Andres, finally, as far as CIC's ownership is concerned, obviously, there's a representative on your board resigned last month. They haven't announced a replacement.
What's your latest understanding of what their plan is for their ownership in AES shares currently?
Andrés Ricardo Gluski Weilert
Sure. Of course we can't speak for CIC, but what I can say is that, first, the decision of Mr.
Zhang [ph] to retire is because he hit the age limit of 70, and in China, they're very strict about the age limit restrictions in government, so that's the first part. The second, as you know, there are things happening in China, so they have the right to assign a new director, and they will do so.
But as you know, because of certain campaigns that are in China today, things have slowed down in terms of rapid decision-making.
Ali Agha - SunTrust Robinson Humphrey, Inc., Research Division
Okay. So no indication from them on what their plans are for their ownership?
Andrés Ricardo Gluski Weilert
That's correct.
Operator
Your next question comes from the line of Stephen Byrd from Morgan Stanley.
Stephen Byrd - Morgan Stanley, Research Division
We saw an interesting development yesterday in Ohio. And I know it's very recent, but it did look, I guess, to us at least, somewhat promising in terms of receptivity to PPAs in the state.
Just curious, I know it's very recent and involves another company, but any immediate reaction to that? Any thoughts as to how that might impact you or just, more broadly, on what you saw yesterday?
Thomas M. O'Flynn
Yes, Stephen. It's Tom.
I'll just hit that briefly. We have been following that.
We did notice that it seemed like PUCO was receptive to the concept of PPAs, but didn't believe that specific example warranted the PPO, or the merits were not significant or sufficiently demonstrated. So we continue to monitor the situation.
We do think -- kind of stepping back on, a bigger picture, we do think capacity will be worth more and that probably was a significant input into us deciding last summer to retain the generation. We are seeing a lot of that, to be honest, through PJM premium capacity product.
That looks like it's on track, so we expect to get more value for our capacity in that fashion. The last thing I'd say, it's a little harder for us.
All of our plants are, let's say, positive on a to-go cost. We don't have any plants that are cash flow-negative, as we see it.
So it's a little harder for us to make the case that, perhaps, some other plants in Ohio might -- the majority of our plants. We do have a small interest in the OVEC plant that might be in a different situation, but the majority of our plants are cash flow positive.
Stephen Byrd - Morgan Stanley, Research Division
Okay. Understood.
That's very helpful. And I wanted to switch over, Andres, to Main Street Power, and as you think about solar for commercial and industrial customers.
I'm just curious how we should think about this in terms of the types of countries that would be most interesting. You talked about the criteria briefly, but is this something that you would envision will start relatively small for some period of time?
Do you see the potential as so positive that within meaningful, relatively short amount of time it could become sizable? How should we think about this in terms of where this goes in terms of scale?
Andrés Ricardo Gluski Weilert
Yes, great question. We're starting off very modestly.
I think as you know, we sold most of our position in utility-scale solar in our JV with Riverstone. We've kept some assets.
We continue to believe that we should have the capability of introducing solar onto our platforms where it makes economic sense. So right now, we're building 20 megawatts of solar in Chile.
So one market that's particularly interesting is Chile, where you have some of the best solar irradiation in the world, and you have very low rainfall in the north, so it's an optimal place for solar. And also energy prices, because they have limited hydro, are quite high.
So that combination of factors makes it an attractive market. We have other markets, such as the Dominican Republic, El Salvador, where, again, you have high solar irradiation and you have high energy prices.
So we feel that this could be a nice addition to our portfolio to address the needs of large customers, especially commercial and industrial. We also think that this is -- distributed solar is something which is affecting the industry, and we want to have our most current information, understand what's happening, understand developments to be able to react faster.
So we think that we have a unique footprint in terms of places where you have the right combinations of natural factors and of market factors, and with this know-how, we're optimistic we will make a good business. Now given AES's size and given that we have a modest investment, at this point, it will take some time to grow out.
And lastly, as you know, we are the world leaders in the use of lithium-ion batteries. We recently won in the Southern California Edison bid.
We got 100 megawatts of more, additional capacity and we're using it really to -- almost like a peaking plant, I would say. So having distributed solar and having the batteries together puts us in a very good position to react to changes in the marketplace with the growth of renewables.
So again, it's part of our having a complete product offering. But again, we're starting off modest and like in batteries, we intend to make money from it.
Stephen Byrd - Morgan Stanley, Research Division
That's very helpful. And if I could, just one last very brief question on Uruguaiana.
You talked, Andres, about the potential restart and the key challenge in terms of getting fuel. What are potential approaches you could take there?
How should we think about the challenges of actually achieving a restart there?
Andrés Ricardo Gluski Weilert
Yes, well, last year it ran for a couple of months and it's going to be running this year as well for a couple of months under emergency situations. Of course, what would be more efficient for us is to receive capacity payments and be able to run it more.
We have, of course, a gas pipeline. It comes through Argentina.
So the issue really has been to deliver LNG into Argentina and get an equivalent amount of gas in Brazil at Uruguaiana. And it really depends on what's the delivered price to the plant, so we've been having relatively high prices of delivered gas to Uruguaiana.
We're working on mechanisms using the existing infrastructure to get lower-priced gas. But what is also very important is that we have been promoting discussions between the governments.
So we had an interconnection line from Chile -- from Argentina to Chile, from TermoAndes into Chile. Given the situation of increased capacity in Northern Chile, we have reversed that transmission line so that we can now export energy from Chile into Argentina.
And there exists the possibility, and that we're exploring, where we would actually pass energy from Chile to Argentina and from Argentina into Brazil. And it would be a win-win for all, and we have had considerable government receptivity.
It's just that we've gotten the permission for the first step, we'll be looking to the second step. But I think what's important, broadly, is that we're looking at ways to decrease our hydro exposure.
And even if rains come back to normal, we don't want to be as susceptible to a drought. Admittedly, this is the worst drought in 100 years but, nonetheless, we want to decrease our portfolio sensitivity to it, so we're looking at these ideas.
Operator
Next question comes from the line of Julien Dumoulin-Smith from UBS.
Julien Dumoulin-Smith - UBS Investment Bank, Research Division
So I wanted to elaborate a little bit on the cost savings. Can you talk about where exactly you're getting from -- them from, and the sustainability as you look forward on the guidance you guys just released?
And then perhaps, the second factor, as you think year-on-year, looking towards the hydrology, what kind of assumptions on normalization are you assuming '15 versus '16, or is that kind of all largely caught up already within the range, the $0.10 range you have?
Andrés Ricardo Gluski Weilert
Okay. Let me take the first one and Tom will take the second one.
The first one, regarding the O&M savings that I mentioned. So first, the G&A savings, the $200 million, that's sustainable.
That's a structural change in how we operate and better use of systems and I think a better organizational structure. So that's, let's say, in the bag.
Regarding the additional $100 million of O&M savings for -- through 2018. Basically, as you know, we're -- have 7,000 additional megawatts, which will come online between now and 2018.
So really, our challenge is to keep O&M constant, flat, and reducing it on a megawatt-hour basis, is what we're going to be doing and also keeping our costs flat relative to local inflations. These are our 2 main mechanisms that we're doing.
Of course, we have opportunities for additional standardization, for additional aggregation of purchases. We now have the systems to do that.
Those are additional things that we'll be looking at.
Thomas M. O'Flynn
Julien, the hydrology for next year, we're assuming it'll be normal, just like we are for Panama this year. Keep in mind that our contract at Tietê of about BRL 200 per megawatt hour ends this year, so we're about 83%, 85% hedged for next year.
So if there was a GSF reduction like we're predicting this year, let's say, 15%, we would lose some sales. We would not be short.
So we'll be in a much better position going forward. And that's how we would expect to run our business at Tietê, as well as everywhere else.
All our businesses that are portfolios in a market we would look to sell some of the power but have some excess that would protect us against reductions in production and hopefully to capitalize on some opportunities where we've got a little bit of length if prices are high.
Julien Dumoulin-Smith - UBS Investment Bank, Research Division
Got you. And then moving to Vietnam real quickly, can you elaborate briefly here on your expectations for not just this year but in subsequent years around earnings trends.
You talked it up in terms of '15 guidance, but in terms of the -- once in service, what does it mean, your latest update on guidance? Because I know you've talked to a lower ROE versus cash flow in that project throughout.
Thomas M. O'Flynn
Yes, so the ROE -- as I said, the ROE will now look closer to the cash yields and the cash yield is in the high teens and the ROE will be in the midteens maybe 14%, 15%.
Andrés Ricardo Gluski Weilert
So, Julien, we expect us to be fully online, both units will be online by the second half of the year. So starting '16, we have a full year of Mong Duong.
Now again, cash is still sort of front loaded. Because it's a BOT, it comes under lease accounting or a version of it, and we have to normalize it over the life of the projects.
Thomas M. O'Flynn
But those are -- Julien, those benefits -- that ROE is sustainable for many years, 10 years plus.
Julien Dumoulin-Smith - UBS Investment Bank, Research Division
Right. But let me just be clear about kind of combining these factors.
You're going to have a better ROE on Vietnam, and so that's another year-over-year benefit in '16, right? So the '15 versus '16 jump is better than what you previously disclosed?
And then in addition to that, in theory, there's a normalization factor that's taking effect '15 versus '16 that should be benefiting as well, right? Is that -- I just want to make sure I'm capturing some of the changes here as you think forward-looking.
Andrés Ricardo Gluski Weilert
The short answer is yes.
Julien Dumoulin-Smith - UBS Investment Bank, Research Division
Got it. Excellent.
And then lastly, just on Bulgaria, if you don't mind my asking. I appreciate that you've taken an impact here.
When you're thinking about the overall value proposition and the NPV, how are you thinking about that in the context of the conversation? And I know that might be sensitive.
I just wanted to get that out there, if I could.
Andrés Ricardo Gluski Weilert
Yes. Of course, we're in the midst of conversations with the government.
For us, the real key point is, one, a prompt payment of the 100% of the outstanding receivables; and the second is really solving the source of the problem, which is the financial situation of NEK, the offtaker. So by really increasing the revenues to NEK, especially on the passing of the C02 taxes, which went to the government.
The government allocated a portion of that to NEK. Now they'll go directly to NEK.
And second, rationalizing the energy that NEK buys because we are the -- among the thermal plants, one of the few EU 16-compliant plants. So if they really enforced their rules, this will put -- make our plant much more important.
And so I think it's -- the exchange, of course, would be that they want a decrease in our energy prices today and for a number of years and, in exchange, we have to look at the overall health of the business. And so we'll be looking at those factors, taking them into consideration.
But I think the most important is that we should end up with a project that we don't have to be talking about receivables next year at this time, and that's really our primary interest in this. And the government, quite frankly, doesn't want to be talking about this either.
But I think that with the cancellation of the South Stream project of gas from Russia, this plant becomes absolutely vital for Bulgaria. So I feel very confident that we will come to a resolution.
Operator
Your next question comes from the line of Maura Shaughnessy from MFS.
Maura A. Shaughnessy - Massachusetts Financial Services Company
Yes, so I just -- I just have 2 questions dealing mainly around Brazil. So just what was the GSF for Tietê in '14 and what is the assumption in '15?
Thomas M. O'Flynn
Yes, Maura, it's Tom. GSF was down 10% in '14, and so that was basically what cost us the $0.07.
Our assumption for '15 is 15% to 17%. Keep in mind, obviously, the prices were a lot lower.
We are filling our short in '14 at basically BRL 700 per megawatt hour. Now the cap is BRL 388.
So it's a larger shortage, but it's cheaper to fill it on a megawatt-hour basis.
Maura A. Shaughnessy - Massachusetts Financial Services Company
And so that the assumption that the GSF would fall into -- and I guess, you're talking about it down from 100%, I think about it the other way. So in the 70s under a rationing scenario?
What's the...
Thomas M. O'Flynn
Well, our -- I guess our assumption in terms of percentage of assured energy, flipping it around, we were 90% of assured last year and we'll be -- we're assuming we'll be 83% to 85% of our assured this year. So the $0.05 that's baked into our guidance assumes that we'll not be rationing.
Obviously, we're watching the rainfall. We'll know more -- I think, at least, the last week or so has been good, but if there were to be rationing, there's different scenarios.
We've looked at them. But our sense is if we were to be rationing, it would be an additional $0.05, so it'll be $0.10 instead of $0.05.
But it's still far too early to call on rationing.
Maura A. Shaughnessy - Massachusetts Financial Services Company
An assured energy level of 75% to 80%, is a guesstimate range on rationing?
Thomas M. O'Flynn
Yes, we have a little more -- well, there's assured but then that also goes into some volume reductions, so we actually start to lose a little bit at Eletropaulo and Sul. So it's a Tietê.
The rationing would be baked in. We'd have some Sul and Eletropaulo sales reductions as well as just Tietê.
That's a Brazil number, not just a Tietê number.
Maura A. Shaughnessy - Massachusetts Financial Services Company
So the $0.08 expected hit for Tietê in '16, I think you had something like 17% of that contract still open. What's the assumption in terms of the pricing to get to that $0.08?
And what are you doing with that open position? What's the strategy there?
Thomas M. O'Flynn
Yes, I'm sorry -- so for -- there is a reduction, let's say, from normal '15 to a normal '16. I want to make sure I separate hydrology from prices.
There is a reduction of about $0.06, $0.07 just on a normal hydro year, from '15 to '16, because this year, we're selling at BRL 200 per megawatt hour. Next year, it's about BRL 135.
It's about BRL 135 while it's been doing a rolling forward hedge program now for about 18 months, and I believe about BRL 135 is our average pricing and about 83% sold for Tietê next year. That number goes up a little bit in the next 3 years.
The reais per megawatt hour goes up a little bit. And the hedging level, just like any rolling hedge program, goes down, about 20% a year.
Andrés Ricardo Gluski Weilert
Exactly -- in -- by '18, it's 46% hedged, and the average price is BRL 141 per megawatt hour.
Maura A. Shaughnessy - Massachusetts Financial Services Company
Okay. And just -- I mean, obviously, the tax is what it is, but when you're selling [indiscernible] forward power today, what would those prices be at today?
Thomas M. O'Flynn
Well, we're seeing very high numbers. If you just did a single '16, we're seeing them in the mid-200s.
But we've generally been selling more -- sorry, go ahead.
Maura A. Shaughnessy - Massachusetts Financial Services Company
Yes, so we know the Eletropaulo contract expires and the like, I'm just trying to understand what that [indiscernible]. The initial power sold was very low prices, but now you have last year the opportunity to reset that cost.
But the pricing hasn't pushed up that much in your asset loss in '16 versus '15. So just wondering what -- is there the chance to be -- is the assumption to be selling power north of [indiscernible] or no?
Thomas M. O'Flynn
Yes -- no, Maura, that's -- I mean, we've been doing a rolling hedge program, as I said, it started probably about 2 years ago. So 1 year ago at this time, we said that we had a fair number of sales for '16 and they were about BRL 150, so as we increased our sales, as we filled up the bucket for '16 over the last 12 months, we've been doing it at about, well, BRL 160, BRL 180, so that's why we -- our weighted average is BRL 135 for '16, which basically reflects sales we've been doing for the last couple years.
It is fair, as we've been doing the sales, Andres has mentioned about 40% sold for '18, we have often sold them in 3-year strips, because there's a little bit of backwardation between '16 and '19, so we've been selling them in 3-year strips to take advantage of the '16 higher prices and basically stretch it out for a couple of years.
Maura A. Shaughnessy - Massachusetts Financial Services Company
So if I were stepping back and looking at what's happening in Brazil, you've had a positive in terms of accounting in terms of Eletropaulo this quarter and on the other [indiscernible] we're supposed to have a special tariff increase in March. So that front for the distribution company, never mind in the next [indiscernible] seeing some rational returns to its [indiscernible] having said that, on [indiscernible] the strategy from the government just seems to be pray for rain.
[indiscernible] Just going to wait until April and that will make a determination or, because it seems easy for them not to have been doing stuff, obviously, a year ago, never mind this year. So what's the -- Andres, what do you think this is going to -- how this is all going to shake out?
Andrés Ricardo Gluski Weilert
Well, I think that -- I remember last year when we talked about this. We said that we thought rationing was very unlikely, given the elections, that they'd do everything possible to avoid rationing.
I think that you'll have the extraordinary tariff increase. We expect that in March.
I do think that will go through, which will help us pass through short-term -- near-term costs. I think that the last couple weeks have been good.
I mean, one of the big problems has been that the rainfall has been falling in the wrong places. It's been falling further south and the big hydros are in the southeast.
So that's what's -- so the average rainfall numbers won't give you the real picture. It's really the fact that they haven't been falling in São Paulo and Mina Gerais, the rain.
So my feeling is that they'll -- they're showing that they're taking active steps to protect the liquidity of the distribution companies. I think they're also taking active steps to run as much thermal as possible.
And so it isn't sort of just hope for the rains. I think they're taking a lot of steps.
So I still think it's more likely that we won't have rationing than we will have rationing. But it's a higher percentage than it was, say, last year.
It might be 40-60 or something like this. We don't know the exact numbers.
But they are doing, taking the right steps to ensure liquidity at the Discos, and also to run the thermals, and they are in discussions with us. At Uruguaiana, we are running the plant and they are really looking at the imports of energy.
I don't know if that sort of gives you sort of our feeling of what will happen.
Operator
Your next question comes from the line of Greg Gordon from Evercore ISI.
Greg Gordon - Evercore ISI, Research Division
Most of my questions have been asked. Good job, guys, navigating through a very tough environment.
One question is when I looked at the shortfall in cash flow that you experienced this year, below the bottom end of the range, how much of that is just stuff that you actually expect to reverse out and come back in '15 and '16 and that's essentially baked into your confidence in the future cash flow?
Thomas M. O'Flynn
Greg, I'd say some of it is, especially the nearest-term stuff will be the Chilean VAT. We've already started to see about 1/2 of that come back.
The Brazilian situation is, as Andres mentioned there have been through the flag mechanisms and there are some meaningful rate increases expected here, whereby the deferred power purchase expense will get passed through here in the next few months. So we have good visibility and we also have a good, reasonable contingency in our numbers.
Greg Gordon - Evercore ISI, Research Division
Great. And when I look at the spread between parent free cash flow and -- sorry, proportional free cash and parent free cash, $1,000,000,000 to $1,350,000,000 going down to $475 million to $575 million, that's a ratio of 42.5% at the low end, of 47.5% at the high end.
Is it a good rule of thumb to assume that proportional free cash translates into the parent free cash at 45% as we go through time or does that rise or fall or is there another way we should think about that?
Andrés Ricardo Gluski Weilert
I think that can rise and fall. I mean, I think we showed it last year, where one fell 25% and the other one grew.
So it will depend on the specific circumstances.
Greg Gordon - Evercore ISI, Research Division
Okay, great. But I mean, should we be concerned that there's that big range of potential outcomes there or is it with a relatively narrow band?
Andrés Ricardo Gluski Weilert
The parent free cash flow, I think we've shown over the years, is very solid.
Greg Gordon - Evercore ISI, Research Division
Okay, great. And my final question is can you just review what's going on with the Southland projects and what is the actual timing of when the capital investments start to translate into net income?
Andrés Ricardo Gluski Weilert
When they'll start to translate? Well, they would come online in 2020, is what we're looking at.
Greg Gordon - Evercore ISI, Research Division
And would you generate AFUDC on those investments or -- in the construction space or no?
Andrés Ricardo Gluski Weilert
No. No, this is...
Operator
Your last question comes from the line of Insoo Kim from RBC Capital Markets.
Insoo Kim - RBC Capital Markets, LLC, Research Division
Just a couple of questions. Regarding the Maritza renegotiations, in your guidance, what percentage decrease in contract prices are you assuming and kind of, I guess, the breakdown of -- from that $0.03, what EPS impact are you assuming?
Andrés Ricardo Gluski Weilert
Well, on the first one, I mean, we can't give any details, because we're in the midst of negotiations right now. So I can't give you any more color on that.
It's in that bundle that we've put up there, of $0.03, and we think that's more than adequate to cover anything that could result from the negotiations.
Insoo Kim - RBC Capital Markets, LLC, Research Division
Okay. And regarding the tax rate for the year, I know it assumes the extension of the CFC look-through rule.
Do you have any updates or timing on when that decision would come out?
Andrés Ricardo Gluski Weilert
We got it last year. Congress has had a tendency to do it last minute, sort of 12:00 on the 31st of December.
So we really don't. I mean, there is a talk of updating the -- starting to make some changes in our tax code, but there remains significant support for this.
And so we expect it to be passed. We're also taking steps to minimize the impact should it not pass, as we have in the past.
So we're now talking about a range of maybe $0.05 or something like that, should it not pass. But just a reminder, we do have $3 billion of NOLs.
We are generating more, so it will always be noncash for the foreseeable future.
Ahmed Pasha
Okay. Well, we thank everyone for joining us today on our call.
As always, the IR team will be available to answer any questions you may have. Thank you, and have a nice day.
Operator
This concludes today's conference call. You may now disconnect.