COPT Defense Properties (CDP) 2018 Q3 法說會逐字稿

Welcome to the Corporate Office Properties Trust third quarter earnings conference call.

As a reminder, today's call is being recorded.

At this time, I'd like to turn the call over to Stephanie Krewson-Kelly, COPT's Vice President of Investor Relations.

Ms. Krewson-Kelly, please go ahead.

Thank you, Kevin.

Good afternoon, and welcome to COPT's conference call to discuss results for the third quarter of 2018.

With me today are Steve Budorick, President and CEO; Paul Adkins, Executive Vice President and COO; and Anthony Mifsud, EVP and CFO.

In addition to the supplemental package and press release related to our results, we have posted slides on the Investors section of our website to accompany management's remarks.

In the press release -- results press release we issued yesterday and on our website, you will find reconciliations to the GAAP and non-GAAP financial measures management discusses.

At the conclusion of management's remarks, we will open the call for questions.

Statements made during this call may be forward-looking within the meaning of the safe harbor of the Private Securities Litigation Reform Act of 1995, and actual results may differ materially due to a variety of risks, uncertainties and other factors.

Please refer to yesterday's press release and our SEC filings for a detailed discussion of forward-looking statements.

With that, I'll hand the call over to Steve.

Thank you, Stephanie, and good afternoon, listeners.

I'm happy to report we had another strong quarter.

We continue to capture leasing associated with strengthening demand throughout our portfolio and remain firmly on track to meeting our profitability goals and modestly exceed our 2018 leasing goals.

FFO per share of $0.50 was in line with the midpoint of our guidance, and we outperformed our guidance for same-property cash NOI growth and occupancy.

The fiscal 2019 defense budget was signed into law on September 28.

That's the first time in 10 years that the DoD received its annual appropriations on time.

As shown on Slide 6, the fiscal '19 base budget came in at $617 billion, representing a 2% increase over 2018 funding and $120 billion or nearly 25% higher than the DoD's 2015 base budget.

An elevated and predictable defense spending environment and consistent bipartisan support in Congress to fund defense are driving demand for expansion space throughout our operating portfolio and for new developments.

As Slide 8 summarizes, our markets are in various stages of growth.

Highlights from last quarter include 161,000 square feet of vacancy leasing in our operating portfolio that, along with our developments placed in service, drove our core portfolio back up to 94% leased.

Notable large leasing gains during the quarter included our Fort Meade, B/W Corridor subsegment, which is up 50 basis points to 92.4% leased; the NoVA Defense/IT subsegment, which is up 80 basis points to 92.1% leased; and our Navy Support group, which gained 160 basis points and is now 93.2% leased.

In terms of development leasing, our third quarter was light with only 19,000 square feet of leasing, primarily at Redstone Gateway in Huntsville, where 2 speculative developments continue to capture the demand from midsized defense contractors looking for more efficient, modern combinations.

Last night, we announced 2 full-building preleases for data center shells, representing the sixth and seventh leases of the 11-building pipeline we outlined for investors last November.

These leases bring our annual development leasing above 1 million square feet, and we expect to meet our upwardly revised development leasing goal for the year.

Regarding our 11-property data shell build-to-suit program, we're under contract to purchase 2 additional land parcels to accommodate the remaining 4 build-to-suits in coming quarters.

Completion of the program is virtually assured at this point.

However, some of our deliveries, and therefore, rent commencements, have been deferred.

The Northern Virginia data center market is the strongest in the country, and demand for suitable development sites is frenzied.

During 2018 alone, we are tracking 16 major land sales in the market to support data center demand, encompassing almost 1,500 raw acres and $750 million of land value.

The sheer volume of data center development activities has stressed the capacity of the county's zoning and power distribution resources, extending the completion times for land development.

This additional friction has extended the completion dates for several of the projects in our 11-building develop program.

While the delayed commencements will defer some of the anticipated 2019 revenue into 2020, the magnitude of the revenue and the value creation of this program has not changed.

I'll conclude my remarks with an update on government demand in our markets.

Demand from government users continues to build, following the increases in the defense budgets.

We completed the 160,000 square-foot NoVA B lease in June and are working to plan the next building at that secure site with a lease execution that could occur in 2019.

We expanded the U.S. Navy again in the third quarter, bringing our U.S. government leasing in Navy Support to more than 100,000 square feet since 2016, when defense budget increases commenced.

We're working on 2 more solutions with the Navy to accommodate additional growth.

And across our portfolio of government demand drivers, we're in discussions with multiple separate government users to accommodate growth or relocation to efficient, secure facilities.

These discussions range from conceptual to advanced planning for near-term and midterm solutions.

Clearly, government demand is rebounding, and we're optimistic about our future opportunities to expand our government relationships.

At 310 NBP, we're in the process of finalizing a lease for the building's second floor and have a high degree of confidence it will be completed by year end.

We continue to work with the government customer to lease the remaining 4 floors, which comprise 120,000 square feet, and currently expect that to occur in 2019.

Each quarter of delay in leasing the remaining 120,000 square feet equates to $0.01 of FFO opportunity cost to next year's results.

Although the timing associated with leasing 310 has been frustrating, we continue to have confidence the property will fully lease.

The 3.3 million square feet we already lease to the U.S. government in 29 full-building leases have been 100% occupied since the original lease dates, some of which go back almost 3 decades.

The long-term value of our U.S. government leases, which generates steadily growing cash flows and require minimal recurring CapEx, far exceeds any short-term implications related to the process delay we've experienced in leasing the 310 asset.

So in summary, the elevated and orderly defense spending environment is spurring new demand from defense contractors and government users.

The DoD's fiscal 2017 budget is driving the year-over-year increases we've achieved in vacancy leasing this year.

Throughout 2019, incremental demand from the DoD's substantial 2018 budget, which passed into law only 7 months ago, should amplify existing demand as defense outlays filter down to the real estate level.

We will likely see incremental demand from the 2019 defense budget emerge toward the end of next year, further expanding leasing opportunities.

We have the right real estate solutions to accommodate expected demand and the appropriate financial flexibility to seize the opportunities as they arise.

With that, I'll hand the call over to Paul.

Thank you, Steve.

Demand continues to accelerate in our markets, as reflected in strong leasing achievements this year.

In the third quarter, our -- we leased 798,000 square feet, including 618,000 square feet of renewals.

And during the first 9 months, we have leased a total of 2.9 million square feet, including 1.8 million square feet of renewals.

For the quarter and the 9-month period, we achieved a tenant retention rate of 77%, which was at the high end of our upwardly revised guidance range.

We have less than 500,000 square feet of expirations left this year, nearly all of which are located in the Fort Meade, B/W Corridor and Navy Support subsegments, where our historical track record has been strong.

Based on discussions with our tenants, we are increasing our full year retention guidance again to a new range of 76% to 80%.

Economics on renewals met expectations.

Cash rents on renewals declined 1.6% in the third quarter and increased 20 basis points in the 9-month period, which were in line with guidance.

Based on deals in process, we continue to forecast cash rents on renewals for the year will be flat to down 2%.

As Slide 11 shows, our leasing CapEx on renewals continues to be one of the lowest among office REITs.

In the third quarter, our CapEx per square foot per year of term on renewals was $1.52; and for the 9 months, it averaged $2.03.

The bigger story for this year is the volume of vacancy leasing that we have achieved, which is depicted on Slide 10.

During the quarter, we leased 161,000 square feet of vacancy, bringing our total for the first 9 months to 348,000 square feet, which is 50% higher than the volume we achieved during the first 3 quarters of last year.

We expect fourth quarter vacancy leasing to be strong, and based on tenant occupancies scheduled, are tightening our year-end same-property occupancy guidance to a new range of 92.25% to 92.75%.

In addition to leasing existing space, our ability to deliver highly leased development projects is critical to achieving our goal of delivering attractive risk-adjusted returns for shareholders.

In the quarter, we placed 214,000 square feet of newly constructed or redeveloped properties into service that were 100% leased, which brought our 9-month total to 450,000 square feet that were 86% leased.

Based on our fourth quarter activity, for the full year, we expect to place approximately 700,000 square feet into service, which will be over 90% leased.

Our shadow development pipeline tracks development leasing and build-to-suit transactions where we believe we have at least a 50% chance of winning, and contained 2 million to 2.5 million square feet last quarter.

Subtracting the 366,000 square feet of data center leases we announced last night, our shadow development pipeline still contains 2 million to 2.5 million square feet.

In terms of composition, traditional Defense/IT office space now comprises 60% of our pipeline and data center shells represents 40%.

Among our 10 active construction projects, 6 projects totaling 950,000 square feet and 1 land parcel collectively represent a total investment of $168 million and are 100% leased.

Leasing activity among the 4 projects that are not yet fully leased is very strong.

At 5801 University Research Court at the University of Maryland's Discovery District, we are close to completing a 10,000 square-foot lease that will fill the building.

At Redstone Gateway, we are close to delivering on 2 spec buildings totaling 79,000 square feet.

During the quarter, we signed an 18,000 square-foot lease with a defense contractor that increased the combined leasing to approximately 50%.

And we are currently in advanced negotiations on 3 leases that will stabilize both buildings by year-end.

Additionally, at 2100 L Street, our trophy development in Washington D.C., we are in negotiations and tracking prospects for more than 1.5x the project's remaining office space.

Given our late 2020 delivery window and the superior quality of our location and design, we feel highly confident in our ability to deliver that asset substantially leased.

More broadly, contractor demand for large prereleases and full-building leases continues to emerge.

We already signed 1 such transaction in the second quarter with our confidential Project EX.

Our shadow development pipeline contains multiple build-to-suit opportunities, and we expect to complete a 50% prelease with a defense contractor on a new office building, if not by year-end, then in early 2019.

With that, I'll turn the call over to Anthony.

Thanks, Paul.

During the quarter, we successfully executed several capital markets transactions.

In late July, we closed on $116 million construction loan for 2100 L Street in Washington D.C. The proceeds from this loan will completely fund the project's remaining development costs.

We continue to improve our balance sheet by match-funding our development investments with equity.

During the quarter, we drew down $80 million of our forward equity agreement and issued $30 million at $30.46 a share under our ATM program.

We'll be using the proceeds from our ATM to fund Project EX, which is incremental to our 2018 development plan.

We now have approximately $86 million of issuance remaining under our forward equity facility and $40 million of capacity under our ATM.

Shortly after the quarter, we entered into a new $800 million credit facility to replace our existing line.

The new credit facility has improved pricing and covenants, matures in March 2023 and has 2 6-month extension options.

Also of note, in September, Fitch Ratings upwardly revised its outlook on our credit rating from stable to positive.

In terms of guidance, as Slide 19 shows, we are maintaining the midpoint of our full year guidance at $2.01 and tightening the range to $2 to $2.02.

We are also tightening our fourth quarter FFO per share guidance from the prior range of $0.48 to $0.52 to a new range of $0.49 to $0.51.

The $0.50 midpoint of our fourth quarter range is flat with third quarter results because we are completing several weather-related R&M projects that were budgeted for earlier in the year.

One last item of note in our fourth quarter guidance relates to a gain on sale we will book on an asset in Central Virginia that we sold last October.

Because we extended a financial guarantee to the buyer as part of that sale, we have listed the asset as held for sale on our balance sheet with an offsetting liability through the third quarter.

On October 1, the guarantee expired, resulting in a $0.01 gain on sale in our fourth quarter EPS guidance.

With that, I'll turn the call back to Steve.

Thank you.

The defense industry recovery is picking up steam, and the timely passage of the fiscal year 2019 defense budget reflects the high level of bipartisan support that exists in Congress to fund defense and intelligence activities.

We've captured new leasing opportunities in our operating portfolio and expect strong leasing momentum going into 2019.

Government and defense contractor demand for new, efficient space is heating up, especially at our Redstone Gateway development in Huntsville.

And demand for data center shells with our customer in Northern Virginia remain strong.

Our releasing gains this year have been primarily driven by the 2017 defense budget.

We're excited at the prospect of new demand that will emerge as the 2018 and 2019 defense outlays work their way through the system and create even more opportunities.

We are uniquely positioned to provide real estate solutions to Defense/IT customers to support their growth, modernization and security requirements.

And with that, operator, please open up the call for questions.

(Operator Instructions) Our first question comes from Manny Korchman with Citi.

Steve, if I'm looking at the presentation you guys put out along with earnings, it looks like on the slide where you have reminders, risks and opportunities, you have a section there that you talk about how long it takes for government appropriations to flow through.

In the deck this quarter, you'd say it takes 9 to 18 months.

In the prior deck, you guys had -- sorry, I had it here a second ago.

You had 6 to 9 months.

So why has that timing sort of, from your view, changed?

Or if it hasn't, why has it changed in the slide deck?

We extended it a little bit in this deck, based on experience we're having with a couple of opportunities that really are flowing out of contracts that got awarded in '16 but are very long-term contracts with major planning components.

And although we expect to get those signed in the next quarter or so, they just took longer than we had previously anticipated.

Smaller contracts continue to be more in the timeframe that you talked about.

Great.

And then turning back to 310, it feels like your comments on this call and then also in that slide presentation make it feel like you have less confidence in that happening this year, and so you talk about more the impact to next year's same-store growth.

Am I reading that right that you don't think that that's going to happen before next year starts?

So we expect 1 floor to get done before next year starts.

And my comments clearly stated that we expect the full leasing of the remaining 4 floors, or 120,000 square feet, to be a 2019 event.

When you say 2019, can you give us more specific -- sort of when in '19?

Is that 1Q, 2Q, end of?

I'm not going to put a timeframe around that.

The situation has been very slow and methodical, and we're just going to be conservative on our expectation on timing guidance.

Our next question comes from Craig Mailman with KeyBanc Capital.

Just want to follow up.

Last quarter, you guys bumped the development leasing target by about 200,000 square feet, and you had mentioned that about 60% of that bump was office, 40% shell.

But with last night's announcement with the 2 additional shells, it looks like almost 100% of that is shells and less is office.

I guess maybe dovetailing with Manny's commentary, just you guys still seem pretty confident about the tail here, but it just continues to take longer and longer.

I mean, is there something else structurally going on with where the demand is going or the amount of space that contractors already have in given markets that is kind of limiting the flow-through on the office side?

No.

What I would say with regard to new development leasing with contractors, major preleases and build-to-suit discussions, they're making decisions for a very long period of time and they're methodical.

As Paul said, we expect to get a major prelease done either by the end of the year or shortly thereafter.

That is with a defense contractor to kick off another building at one of our government demand-driver locations.

And we have several others that we're in advanced planning on.

So we're confident we'll get them.

Okay.

And then just -- I think, Anthony, you had mentioned that the ATM equity is being used to fund EX.

I know you guys are limited in what you can say there, but can you kind of talk about the decision to issue on the ATM at, I guess, I estimate it at about a 7% cap?

Given your data center developments are low 7s, your office are 8% to 9%, kind of where does EX fall in that spectrum of returns?

And can you just talk about the appetite to continue to issue more below NAV to fund the project?

And the timeframe to kind of make up that dilution as you get projects done?

So the Project EX's initial cash return is probably in the low 8% range from a cash NOI perspective and then has annual bumps to it.

With respect to where we issue under the ATM, we're looking at that, funding that project or funding other data center shells or office -- defense contractor office investments, and looking at that capital allocation decision not just against the current -- the initial cash yield, but also against the development premium that we're creating with that investment.

So when you look at that 8% or 8.25% yield -- initial cash yield from Project EX, we would argue that's a 5% to 5.25% asset, given the tenancy and the mission that it's connected with.

So the price, the slight discount that we're issuing the equity at relative to the development profit that's being created is a relatively small component of that NAV creation.

So we're very judicious in terms of the price that we're willing to execute under the ATM, but we're keeping in mind that the value we're creating with that -- those development projects is not at its initial cash yield, it's at a cap rate that, in most cases, is several hundred basis points below that.

Okay.

And then I know, Steve, we talked about in the past, joint venturing more data center shells.

Just given the success you guys had on leasing there, the thought process on maybe putting more in that JV.

And kind of just maybe some color around GI's appetite at this point.

Are they still there to do more with if you guys wanted to?

So let me answer the first part of the question.

We've said before and continue to maintain that when we hit a situation where we need internal equity to fund new development, that would be our lowest-cost source of capital.

With regard to GI, we haven't really had a conversation with them in recent quarters.

But when we have had conversations, they continue to have interest.

But I have to tell you, we get unsolicited calls regularly from investors looking to get into data center investment.

And I'm quite certain, if we were to seek an additional JV, we'd have multiple choices of people that would be willing to get into that structure.

Do you have a sense of what pricing would be for those assets today?

I fully believe it will be at, plus or minus, 5% cap.

Our next question comes from Jamie Feldman with Bank of America Merrill Lynch.

So I was hoping you can talk more about the delayed land in Northern Virginia.

Just what do you think the earnings impact would be on that?

And is there a chance that your tenant could actually walk away if you don't get it done by a certain time?

So the first of part of that, we're not ready to calibrate that just yet, and we will be as we move into guidance.

We just wanted to elevate awareness that there's some timing delay.

With regard to the confidence that the customer will be there for us, 100% confidence.

And it's just complications in getting the land development done so we can kick them off.

Okay.

So maybe just provide -- if you could provide more color around it.

So when did you think you'd have it done versus what's reasonable now to be able to close and start (inaudible) and finish construction?

So there's 2 separate land parcels.

One of those, I think, we handicapped for an August or July closing.

That one will happen near December.

And then a second, we'll close in the first quarter due to some complications in getting certain things that we need to develop on that land.

And so -- and then how long would it take to actually get the building up and running?

Is it 6 months?

Well, each site will accommodate 2 buildings.

So the first building I would guess to be about 6 months, and the second building, 3 to 4 months after that.

Okay.

Anthony, I guess, just you've talked about -- well, then also 310 NBP.

I mean, what -- is there anything structurally or procedurally that would delay you from getting it done by a certain time next year, that last 4 floors?

So the government is working through the steps they need to get the authorizations that they need to lease space.

And that timing is driven by them, not us.

Okay.

But there's no, like, known dates that it can't happen before?

It's more just a matter of them getting their authorizations in time?

That's correct.

Yes, waiting [and everything].

Okay.

And I know you guys haven't given '19 guidance, but it does seem like there's some moving pieces here.

Are you able to just walk through some of the big blocks, just in terms of where we know there will be some growth and where we know there will be some drag, including lease accounting?

I'm sure.

I mean, in terms of where there will be some growth, there will be some growth from the developer projects that we placed in service, in addition -- including the incremental projects to our pipeline that we had anticipated at the beginning of the year.

There will be some benefits from the leasing velocity that we're experiencing in the existing portfolio.

That's a benefit from the 2017 defense budget.

And that will be partially offset by the risks that we outlined in the current deck, the timing of 310 as well as the delay in some of the deliveries of the data center shells.

With respect to the changes in accounting, the impact for next year's earnings of the change in lease accounting is a little over $0.01 a share.

Okay.

And then finally, you guys have made good progress leasing up vacancy.

Can you just talk about, if you look at the portfolio today and the desirability of the space or -- I guess maybe a better way is to compare the pipeline you're seeing of demand versus what you actually have available and how does it line up?

Like, kind of what's the shadow pipeline for your vacancy right now as opposed to just your development pipeline?

Jamie, it's Paul.

Well, first, we're certainly quite pleased with the progress we've made so far in leasing velocity.

Each quarter has increased.

And as we said, we're 50% over where we were last year during -- over the same period in terms of leasing velocity.

We see it continuing through the fourth quarter and expect it to continue into next year.

But the alignment of our vacancy with the activity is satisfactory.

It's not perfect, but it's satisfactory.

And so we have at least as many prospects for the spaces we do have vacancy, which is right around 1 million square feet.

So we keep having the activity that matches well with our vacancy.

Let me just throw some numbers around that, Jamie.

What we call pipeline or advanced negotiations, those are deals that pipeline is 50%-or-more likely to close.

We're at about 300,000 square feet in pipeline or better.

If you add prospects, where we're working on solutions, it's about 600,000 square feet.

And in total, the full breadth of opportunities identified is nearly 1 million.

So we've got really good volume behind it.

So you're saying 300,000 square feet is just for vacant space.

That's all vacant.

Yes, that entire report deals with the operating portfolio.

Our next question comes from Tom Catherwood with BTIG.

Just wanted to focus on the data center shells a little bit more here.

When you laid out the original 11 opportunities back in October of 2017, the average cost was something in the neighborhood of, like, $190-plus a square foot.

The last few that you started have been in that $160-something a square foot kind of range.

The 2, though, you announced yesterday are like $125 and $137 a square foot.

With pricing pressures and land costs, it was kind of a surprise.

What's keeping those prices so low?

So the 2 that we announced yesterday are in a land parcel that we already owned, so we had a much better cost basis in the land.

And really, those are being developed at that location because another site, where we had tried to acquire, we could not get the power in at the timeframe that met our requirements.

So it's just a fortunate opportunity to pass on a nice, efficient cost basis to our customer.

And Tom, just to give you the totals, our original 11 data center pipeline had a total of $383 million worth of development investment.

Our current budget for that, for the 11 projects, is about $375 million.

So it's slightly less.

But some of the ones that will be coming are going to have a higher burden of land cost.

Got it.

So that kind of leads into the next part of my question, which is, so the spend has been a little slower than expected, partly because of acquiring land, partly because of being able to do it on land with a lower basis.

As of last quarter, I think the expectation was that the forward equity could fund development spending through the first quarter of '19.

Given slower spending and the ATM issuance, what are your expectations as far as how far your equity can now fund what you have in the pipeline?

So it will take us into the -- in terms of what the spend is, it will take us most likely through the end of the second quarter of next year.

But there is a firm funding date under the contract we executed to fund those proceeds by the end of April of 2019.

So there will be a bit of a disconnect in terms of timing because of the 18-month length of the forwards.

Got it.

Okay.

And then finally for either Paul or for Steve on the shadow pipeline, my math might be off here.

But if I kind of pull out the remaining data centers, it looks like you have anywhere from kind of an additional 1 million to 1.7 million of kind of opportunity beyond the known data center shells.

A, am I in ballpark there?

And then b, what's breakdown as far as additional data center shells versus kind of traditional office-type opportunities in that other bucket?

So your math is pretty right on the money.

There's about 900,000 square feet of data center shell in that 2 million to 2.5 million.

So the non-data center shell component is 1.1 million to, call it, 1.5 million or 1.6 million square feet.

And the second part of the question was?

No, I was just trying to figure out if, with the balance of that 1.1 million to 1.6 million, was it all office?

Was it some additional data center shell beyond the original 11?

And it sounds like it is just office.

It's primarily -- yes, it's substantially office.

It's all office.

Our next question comes from John Guinee with Stifel.

Great.

First, it's been 1 year since you did the forward equity sale.

Great execution.

Congratulations.

Two sort of subject matter.

First, Steve, I agree 100% that the AWS powered shell exit cap rate's about a 5%.

We've been told that -- from your competitors, that you've got to build a little bit less than a 6% yield on cost to win that bid if you're going to exit it at a 5%.

And then so comment on that.

And then I guess the second question would be, you've got 4 more of these powered shells that you don't own the land yet.

How have you structured your deal with your client, given that your costs have to have gone up 10%, 20% in the last year just because land prices have doubled out there?

So first answer, we've never been -- we've never done a deal sub-6% on our development cost.

And we're in the high 6s today.

Second answer, we do our deals as a return on cost.

So in essence, increased cost in land doesn't affect our initial yield.

Got you.

Okay.

And then I guess the second comment would be, you're sitting there with 310 NBP.

BXP has a full building at Annapolis Junction, Emerson Corporate Commons has a few floors vacant.

Those are the 3 primary secure space available for the NSA or the Department of Defense to lease.

At the same time, as near as I can tell, there's about 1 million square feet of space, office space, under construction on Fort Meade.

You got JOCC, Joint Operation Cyber Command, for 240,000 feet.

Marine Corps Cyber Command looks like about 120,000, 140,000 feet.

There's another 600,000 square foot building under construction.

Are you sure that the DoD budget isn't oriented towards military construction, or mil con, rather than leasing space?

Well, there's no question, a lot of mil con activity.

But as we've explained before, that's a recapitalization of the campus facilities on Fort Meade.

If you look at the facilities that exist, none of them meet anti-terrorism force protection standards.

And some of the newer buildings are early '80s construction.

So yes, we're confident that what I've conveyed to you, that the demand on Fort Meade will lease our building.

It seems to me they might be putting a lot of money in these buildings that you think are going to be torn down soon.

But you guys are the experts.

Our next question comes from Jason Green with Evercore.

Just a question on what you mentioned earlier with the 2019 budget passing so early.

Is it fair to say that the impact from that budget will be felt in your fiscal year '19?

And does that materially increase expectations for the year?

Yes, exactly.

So we expect the fiscal '18 budget, which has now been passed for 7 months, begin to ramp up into '19.

And the fiscal '19 budget, which just passed last month, will start ramping up near the third and fourth quarter of 2019.

So essentially, it will be a unique year, where you kind of are feeling the impacts of 2 budget proposals?

It will be.

Great.

And then just one question for you.

With the dramatic increase in budget proposals, have there been new entrants into the market, new kind of real estate providers for defense contractors that recognize this as an opportunity?

None.

No.

Not really.

Our next question comes from Rich Anderson with Mizuho Securities.

So I would describe your business as exciting but complicated and time-consuming, and particularly as it relates to the defense budget you just described.

The portfolio is built for 2.5-plus same-store growth.

I'm wondering, when you consider some of the timing behind even having a sort of a double whammy positive in 2019 from 2 budgets, does the recovery back to sort of a more normalized internal growth rate happen in the hockey-stick fashion?

Or do you think it's more of a gradual return over a period of a few years?

I think it's just gradual step.

Because, Rich, that -- the benefit of that leasing activity, the 2 budgets combined, will be to leasing in 2019.

And leasing will have its benefits to same-office cash beginning late '19 into '20.

So it's really the double-up of the budgets in the late third and fourth quarter of next year that will really impact leasing that will then impact subsequent years.

Okay, good.

Steve, you had a nice pecking-order comment there about the leasing opportunity, 300,000 in the current portfolio, 300,000 of new development and then perhaps 400,000 longer term.

When you think about some of the discussions that you talked about having from conceptual to advanced planning, to what degree are you kind of pushing them into the top category in the interest of getting the first pecking order done and also seeing the impact on your cash flow more quickly?

Or are you ambivalent to that?

Are you happy to have them do a full new development, even though it might take longer to see the fruits of your labor?

So first, I want to just clarify a little bit.

The 300,000, 300,000, 400,000, that has to do with the progression of lease opportunities only in the operating portfolio.

And then [working] to the conceptual to advanced planning, that has to do with planning new facilities for multiple government users at multiple locations.

And there's no inventory that exists that I could push them into.

Got you, okay.

And then -- I thought I got that wrong, by the way.

So sometimes, you have to ask a stupid question to get a right answer.

The last question I have is, let's assume for the moment that Amazon is on its way to the region.

If you were to apply your own crystal ball, do you see the company changing a bit in response to that potential demand?

In other words, doing more in the way of more conventional office investing, development or acquisitions, in the area?

Or will it not impact your sort of strategic approach to government and all that stuff?

Well, first of all, my crystal ball is not very good with regarding Amazon HQ2.

Flip a coin, it's your guess.

We would certainly welcome them picking any of the locations they've considered in our marketplace.

And I think many boats would rise with the tide that, that would bring to us.

With regard to our strategy, there will be no change in our strategy.

We are laser-focused on our defense, data center shell, which is with cloud computing providers to the government.

We have a bit of Regional Office.

We see value-creation opportunities in those, and we'll continue to stay laser-focused on what we laid out.

Right.

Okay.

And then last question on the Regional Office.

I mean, is -- could that ultimately become a source of funds for you as you maybe go full bore into strategic assets in light of where your stock trades today?

Absolutely.

It would be our second priority of internal capital sourcing.

But having said that, we have value-creation opportunities in each one of those buildings that we're pretty focused on realizing before we harvest it.

So we'd prefer a couple of years down the line.

Our next question comes from Dave Rodgers with Baird.

Maybe to tighten it a little bit with Rich's question.

But when you talk about kind of small tenants being more visible and maybe not really having a lot of places to push them into from a strategic standpoint, do you anticipate having to?

Or would you be comfortable moving to a higher degree of speculative construction through '19 and '20, given where the budget is and just having space to meet demand pretty quickly?

So it's market specific.

And in most of our locations, we've talked about our occupancies in that 92% to 93% range.

We have capacity for small and medium tenants.

The one location where we moved to a more speculative development was Huntsville, where we kicked off 2 buildings, roughly 8,000 feet.

We're pretty sure they will be stabilized by year-end.

We -- I would not be surprised if we continued building to meet the demand that we see down there because it's pretty exciting.

So if you see a spec from us, it'll be at Huntsville.

We're also planning another building at the University of Maryland, but we expect to have a prelease associated with that.

Right.

And then maybe just a final wrap-up on the 1 million to 1.5 million of office pipeline that you talked about.

You were pretty clear on your returns for the data shell business, and that was helpful, so thanks for that, Steve.

What are you anticipating over the next couple of years in kind of the market level for this office pipeline?

And can you hold something north of an 8%?

Or are you seeing some incremental pressure on returns there?

We think we would be 8% or better.

On average.

And then lastly, maybe on timing, and maybe that's for Paul, talk about the timing for that office component.

It seems like the data shell component is a little bit more clear, office a little bit more cloudy.

But can you kind of give any goalposts around the timing on that longer-term office component of the pipeline?

Well, again, it's sort of market-specific, because as we said in our script delivery, some of them are near-term, some of them are long-term evolutions.

But I think in '19, we will see a number of those projects come and advance to complete in '19 and '20 in terms of the office projects and then signing them up.

So each market kind of has their own time frame that they -- that we're working with and involves government uses as well as defense contractor.

With regard to defense contractors, we'd like to get one announced by year-end, but may dribble into January.

Our next question comes from Chris Lucas with Capital One.

Steve, on the data shell side, have you built any two-story projects for your client?

We're in the process of building several.

Okay.

Is the stuff that you've already built capable of going 2 floors?

No.

That would not be an efficient conversion.

And have you delivered 2 floors to them at this point?

We have not.

And I'm not showing any further questions at this time.

I'd like to turn the call back over to Mr. Budorick for closing remarks.

Thank you all for joining our call today.

We're in our offices this afternoon, so please coordinate through Stephanie if you'd like a follow-up phone call.

Have a great day.

Thank you for your participation in today's Corporate Office Properties Trust third quarter earnings conference call.

This concludes the presentation.

You may now disconnect.

Good day.