Good day, ladies and gentlemen, and welcome to the Third Quarter 2012 American Public Education Inc. Earnings Conference Call. My name is Seanteley, and I will be your facilitator for today's call. [Operator Instructions] As a reminder, this conference is being recorded for replay purposes. I would now like to turn the call over to your host for today, Mr. Chris Symanoskie. Please proceed, sir.

Thank you, operator. Good morning, everyone, and welcome to American Public Education’s Third Quarter 2012 Earnings Conference Call. Presentation materials for today’s call are available in the webcast section of our Investor Relations website, and are included as an exhibit to our current report on Form 8-K filed earlier today. Please note that statements made in this conference call regarding American Public Education or its subsidiaries that are not historical facts are forward-looking statements based on current expectations, assumptions, estimates and projections about American Public Education and the industry. These forward-looking statements are subject to risks and uncertainties that could cause actual future events or results to differ materially from such statements. Forward-looking statements can be identified by words such as anticipate, believe, could, estimate, expect, intend, may, should, will and would. These forward-looking statements include, without limitation, statements about the fourth quarter 2012 and the full year 2012 as well as other statements regarding expected future growth. Actual future results could differ materially from those expressed or implied by these forward-looking statements as a result of various factors including the risk factors described in the Risk Factors section and elsewhere in the company’s annual report on Form 10-K filed with the SEC, the company’s quarterly reports on Form 10-Q filed with the SEC and the company’s other SEC filings. The Company undertakes no obligation to update publicly any forward-looking statements for any reason, even if new information becomes available or other events occur in the future. This morning, it's my pleasure to introduce Dr. Wallace Boston, our President and CEO; and Harry Wilkins, our Executive Vice President and Chief Financial Officer. Now at this time, I'll turn the call over to Dr. Boston.

Good morning, everyone. In today's conference call, I will review the results of our third quarter operations, highlight recent academic successes and discuss our long-term strategy. Then Harry Wilkins, our Chief Financial Officer, will discuss our financial results in more detail and provide additional perspective on our outlook for the fourth quarter of 2012. For the 3 months ended September 30, 2012, overall net course registrations increased 17%, compared to the prior year period, and net course registrations by new students were approximately flat year-over-year. Net course registration growth was in line with our expectations, and the results are especially noteworthy given the measures we implemented to reduce financial aid fraud and abuse and in light of the fact that our prior year comparisons may have included a high percentage of so-called course takers. During the third quarter of 2012, net course registrations by students using Department of Defense tuition assistance, or TA, increased 10% year-over-year as a result of new enrollment growth and improved persistence of active-duty military students. This growth is rather significant given the already sizable military population at AMU and the budgetary constraints possibly facing the U.S. military. During the third quarter of 2012, net course registrations by students using veterans benefits, or VA, increased 93% year-over-year. We believe that this increase is largely related to the addition of a partial housing allowance for post-911 GI Bill students attending distance learning programs that was effective October 1, 2011. Prior to that date, no housing allowance was permitted for students who attended distance learning programs. Moreover, the company is not aware of any significant proposals to modify future veterans GI Bill benefits. During the third quarter, net course registrations by students using cash and other sources increased 7% year-over-year. This growth was propelled in part by expanding…

Thanks, Wally. Turning to Slide 7. American Public Education's third quarter financial results include an 18% increase in revenues to $77.1 million, compared to the third quarter of 2011. The revenue increase was primarily driven by growth in net course registrations from civilian, military and veteran students. Operating income for the third quarter 2012 increased 17% to approximately $17.5 million. General and administrative expenses increased as a percentage of revenue to 20.8% from 18.6% from the prior period -- prior-year period. However, the increase was lower than expected due to improvements in bad debt expense, resulting from less FSA abuse. Bad debt was approximately 4.4% of revenue during the third quarter of 2012. While bad debt was up from 1.7% of revenue in the third quarter of 2011, it was sequentially down from 5.1% in the previous quarter of this year. Instructional costs and services decreased to 34.2% of revenue in the third quarter of 2012, compared to 36.7% in the third quarter of 2011. This decrease was primarily related to cost savings from our ePress initiative, as well as efficiencies gained from process improvements in our student support services. Selling and promotional expenses as a percentage of revenue increased to 18.7% of revenue, compared to 17.9% in the prior-year period. The increase was lower than expected due to the timing of certain marketing expenditures, whereby approximately $870,000 of expenses that were originally budgeted for the third quarter are now expected to be incurred in the fourth quarter instead. The company believes that S&P expenses as a percent of revenue should remain at or below 20% for the full year 2012. In the third quarter of 2012, net income was approximately $10.8 million or $0.60 per diluted share ahead of guidance. Please also keep in mind that the prior year…

[Operator Instructions] Your first question comes from the line of Jerry Herman of Stifel, Nicolaus. Jerry R. Herman - Stifel, Nicolaus & Co., Inc., Research Division: I guess the -- maybe the first question, if we could just talk about the corporate relationships. Can you guys give us an update on the metrics there in terms of how many there are? And as part b, it looks like Walmart might be getting a little bit more visible in terms of the internal advertising of that program? Is that fair?

Let me try to start with an estimate. We have relationships, Jerry, with somewhere in the neighborhood of 50 to 60 corporations. And none of them other than the Walmart relationship is -- towards Lifelong Learning Program, is exclusive. So we're one of several, typically, vendors who provide education services to their employees and in a few cases we actually do direct billing with those corporations. As far as visibility with Walmart, we -- the answer is yes and no. We just mapped some of our courses in organizational development to some of their management expectations for development as well. And that particular program, 100% of the tuition will be paid directly by Walmart's HR department. So that's a little different. That's something new. The existing program, we renewed for 3 years, and we expect that after the busy season's over, we're currently in their holiday season, preferably on February 1, is sort of a tracking date, they will once again ramp up marketing the Lifelong Learning Program. Jerry R. Herman - Stifel, Nicolaus & Co., Inc., Research Division: Okay. And then just a quick follow-up with regard to sort of the competitive landscape. A lot of the for-profit players are reducing or incentivizing scholarship-ing granting money to their students. Do you guys -- I know this is a tough one, but do you guys see any influence of that in your student interest?

We don't see any negative influence. Our scholarships vis-à-vis our corporate relationships aggregate 1% of revenue. We prefer not to give any scholarship, but faced with a request from a strong and large company, we will give something. And it's currently 1%. It's pretty stable. I think that the discounts that the other companies have offered, at least at this point in time, have not negatively influenced that enrollment.

And, Jerry, just to go through it one more time. We believe that if you took out the abuse students last year on a real-comparative basis, we think our guidance for the fourth quarter this year would've been about 8% or 9% new course registration growth and about 19% to 20% net course registration growth, if you excluded those abuse students last year. So we think that we haven't really seen much impact from the competition with reduced prices.

Your next question comes from the line of James Samford of Citigroup.

Just a quick follow-up on that. I think you've installed the $50 tech fee at this point. I was wondering if -- is there any negative impact on new starts or retention from that at this point?

We're not aware of it. I think it's a little too soon to see, James. We had for basically 2 reasons. One, to ease our students into the idea of paying a fee, and also for the ease of programming it with our computer guys. We announced -- When we announced the fee, we said that if you register before September 10, we would allow you to take that class without the fee. So people can register for up to 5 months worth of courses and we're sure that some students who are being pretty frugal, as many of our students are, registered for months out. So we haven't seen the true positive impact of the fee yet, and we haven't seen a negative, at least with the students who might not want to pay the fee because they were able to register for 5 months out. So, I can tell you, though, whenever we have -- we've announced 2 increases in graduate tuition over the last 10 years, since I've been either CFO or CEO. And this particular time period, I only received 2 negative complaint letters. So I think people were pretty open to the fact that it's been such a long time since we've had an extra fee or something that there just weren't that many complaints.

Yes. And to be honest with you, we really believe that allowing the students to register in advance of the technology increase, which we implemented that first week of September, probably had a -- we probably would have had about $1 million more revenue in this period had we not let them preregister for the those classes without the tech fee. So that -- I don't know whether people put that in their models or not, but that was the impact of allowing people to register early.

And just as a follow-up. You got quite a range of opportunities right now between ePress, your corporate partnerships, it sounds like some international and even an outsourcing opportunity that you're looking at. I was wondering, can you prioritize the sort of near-term impact of some of those? And I guess, really, do have the -- where are you on an organizational level to pursue a lot of these opportunities at this point?

Well, I'll speak to 2 of them and let Harry talk about the others. On the ePress, we're still tracking. Our goal was to get $8 million in savings for the full year in 2013, we have no information, at the least, that we won't. And then on the international opportunities, we've been talking about that international market publicly in our quarterly statements for over a year now. We've been thinking about it for 2 or 3 years and our research really indicated that to be effective at that, we needed to have, as the military says, boots on the ground in those countries. And we looked at the cost of -- even if we had a person who was capable of covering several countries, the cost of doing that and funding it particularly in the startup situation versus partnering with New Horizons. And we ultimately concluded that with their multiple locations in a bunch of countries, where they truly have boots on the ground, that it was much more cost-effective and probably likely to lead to a greater and quicker benefit. So we were pleased to be able to go from, really, a relationship a few years ago with them where we were evaluating some of their programs for credit to this partnership and investment. Harry, do you want to talk about any of the other questions there?

No. I mean, we were -- we're very excited about this relationship with New Horizons. They have some great corporate long-term relationships with companies that we have not historically dealt with. We think our corporate outreach team can work closely with their corporate outreach team to put together a really good corporate training and educational program for high-tech professionals, not just domestically, but worldwide. So this is a great way to introduce our -- we would've had a difficult time on our own sitting in Charles Town, West Virginia to effectively get the word out about our programs internationally. We feel with working with New Horizons network, they can put us in touch with working adult professionals in 63 countries. We may want to take advantage of the fact that some of the New Horizons courses, the high-tech courses that they're getting training in, may or may not convert to credit in our program. But then we have -- they'll get the word out about our program, similarly the way Walmart gets the word out about our program through their employees. So this is -- it goes right along with our philosophy of relationship marketing, partnering with high-quality companies and helping to educate professionals. So we're excited about it. We think it will have a very good long-term impact.

Your next question comes from the line of Trace Urdan of Wells Fargo Securities.

Just continuing on that line. Can you describe the nature of the financial relationship? Are they going to be actively marketing your degree programs internationally? Are the students going to be taking the courses physically in their centers on their machines? Can you give us a little bit more detail on how that's going to work?

Trace, we don't have an agreement like that. That's a great idea. Our focus was to do 2 things initially. One, to have our academics separately and independently evaluate the updated programs that New Horizons offered. Since the last time we had done it, they've added some additional certifications in VMware, for example. And to make sure that that was done appropriately, and they're wrapping that review up. The second thing was that we would look into ways that we can make that agreement known at the multiple international locations and how we choose to do it hasn't been cemented or negotiated at this time. But we certainly think there are a number of venues from simply having brochures available like we do in military educational service offices at military bases around the world, to perhaps some more direct relationship. But as you may know, international marketing relationships are governed differently than U.S. marketing relationships. So whatever we would do internationally, we would more than likely not do the same in the U.S.

Your next question comes from the line of Corey Greendale of First Analysis.

So your net [ph] should give us an estimate of what the new registration and total registration growth would've been in Q4 x the stipend chasers. Do you have those numbers for Q3 as well?

I think we gave them on the last call, Corey. It's in the transcript from the last call. We don't have it sitting here with our notes today.

Where I'm coming from is that the comps was equally tough in Q3 versus Q4. So I'm just trying to get a sense if there was kind of an underlying slowing of the growth rate? And if there was -- what you think that's attributable to?

Well, actually, the help was -- it's a tougher comp in Q4. Q4 was the -- I guess, last year was the high point of our FSA abuse students. We actually have a tougher comp in the fourth quarter. But again, it is what it is. I think that our guidance, as always, is conservative but good.

Yes, I think -- and you might note that we -- the guidance we gave last quarter on comparative, as well as this quarter, was based on analysis of course failure rates. So at that time, a year ago, when we were reporting that we had this influx of the students and the problems, we really couldn't identify exactly to what extent it was until we saw the ultimate outcome of the courses, which the change, the delta and the upgrades between what we typically expect and what we saw, because basically once these folks got their refund checks, they walked and get an F in the course, is what we used to do the analysis, retrospectively.

Yes, and I think you're seeing in the fourth quarter and then for the full year of 2012 now, it's becoming clear the double whammy impact this year that we got from those abuse students. We have the tough comparative number on the top line and then the bad debt expense. It looks to us like our bad debt expense for the first 9 months is up $6 million over last year. The majority of that is due to those abuse students. Well, that $6 million is the equivalent of $0.20 EPS of bad debt expense that we had to run through this year related to those -- the impact of those abuse students last year. So I think that we're showing that as those students have gone out now and as we're getting back to more normalized operations, we saw in the quarter our bad debt expense came down from 5.1% last quarter to 4.4% this quarter. I think we're starting to get out of that. The tough comp of last year really had the double whammy of tough comp of -- top line growth and then additional expense this year of bad debt. Christopher Shutler - William Blair & Company L.L.C., Research Division: Okay. And then on the Walmart relationship. So if you just look at the renewals of the existing relationship, the Lifelong Learning, has Walmart indicated whether they're going to start providing tuition assistance? And then on the newer, the professional development opportunity, can you just give us a sense, will those people who enrolled be included in your registration numbers? And what the price level will be for that?

Yes, we'll -- obviously, to give out current numbers from Walmart, we'll have to get their permission. But, yes, that new program, we'll keep them in there, in the numbers that we report. And as far as timing of a reimbursement program for the Lifelong Learning, that -- the choice in that is up to Walmart. We believe that when they plan to start marketing again the 1st of February, that there will be some announcements, but it's not ours to discuss. So...

Your next question comes from the line of Peter Appert of Piper Jaffray.

So the margin compression you've seen thus far in 2012 obviously in the context of some of these new initiatives you've got going on and the bad debt that you referenced, the guidance suggests it continues in the third -- or excuse me, in the fourth quarter and maybe at a somewhat higher rate than we saw in the fourth quarter. So Harry, can you help me understand if there's anything different going on in terms of expenses? And then more importantly, I guess, how should we think about 2013? Do you think margins have stabilized at this point?

Yes, we're, as we have said, I think, on the previous call, looking forward, even though we're not giving any guidance at this point for 2013, we would expect full year impact of our ePress initiative, which we think is going to be about a $4 million savings in this year. We think it'll be more like an $8 million savings, next year. We'll see the impact of that. We'll see the impact of some of the expenses we incurred this year to prevent abuse students from getting into our system, we think will create efficiencies in processing financial aid next year, which we should start seeing particularly in the latter half of next year. I think we'll see even more as we implement a few more things to streamline that process, which will get our cost down, and I also think our bad debt expense will return to a more normalized level. And the other thing we had talked about on the call that nobody's asked about is the marketing spend. We did move about $870,000 of marketing costs, which we had originally planned to spend in the third quarter into the fourth quarter this year and that just the normal timing of -- we watch how normal trends are going, and we watch our marketing spend, and we decided to do that in then -- at the -- in September. So I certainly think that we should start to see some margin improvement next year.

Okay, great. That's very helpful. And Harry, how does the tech fee then flow-through in terms of revenue per student? You didn't see it this quarter, which I understand because of the preregistration. But we should start to see that in terms of some modest increase in the revenue per student in 4Q, correct?

Yes, that is correct and certainly, in next year. You'll see it in the back end of the fourth quarter this year and then at the beginning of the first quarter of next year. It equates to about a 5% increase for roughly half of our students.

Your next question comes from the line of Jeff Meuler of Baird. Jeffrey P. Meuler - Robert W. Baird & Co. Incorporated, Research Division: Just wanted to ask a follow-up on Corey's question around the 8% to 9% underlying starts growth guidance for Q4. I think for Q3, you guys had said mid-teens -- I guess, the technical question on the definition of net course registrations for new students. If a student would register on, say, September 5 to get ahead of the September 10 cut-off date for the technology fee but would be registering on September 5 for a course that starts on October or November, would that student count as a net course registration from new students in Q3 or in Q4?

Q4.

It will count in Q4. And again, they the only count if they make it to the add/drop period. So just registering isn't enough. You actually have to show up for class and make it through add/drop before we count you. Jeffrey P. Meuler - Robert W. Baird & Co. Incorporated, Research Division: Okay. But the difference between the mid-teens underlying in Q3 and the, I guess, 8% to 9% in Q4, we shouldn't take that, that you're seeing any signs of incremental underlying softening, correct?

No, I mean, we're trying to give guidance in a period -- we have a little bit of uncertainty, and we didn't want to bring it up, really. I'll bring it up now because it's -- you've asked the question, but there's a little bit of uncertainty about the impact of the Hurricane Sandy. We have seen about 650 students who have been unable to get their books. We're in the first week of classes in November. The number of people who've been able to log in to the classroom is so far -- they have to log in to a classroom by Sunday night, midnight or they'll be dropped. The number of people compared to prior months is less. In other words, we believe that there may be a portion of the country where students are unable to log in and that may impact. So we tried to built that into our projections. You also have to remember that a lot -- we have a lot of students who are first responders, and as we've seen in the past when Hurricane Katrina hit, then we had the -- when the oil spill in the Gulf of Mexico happened and we had full deployment of the Coast Guard, when we've had periods where we've had national disasters in our country, it has had a short-term impact for a month or 2 on our enrollment in the past because we have so many students who respond to those situations.

But I would add that we aren't sitting back and just saying, "Well, what happens, happens." We tried to contact all 650 students who were impacted by Sandy. And by impacted, we were notified by UPS and FedEx that they couldn't get deliveries to them to make them aware that we would put e-books in the classroom for them, as well as we would have liberal start guidance for our emergency responders, particularly some people who actually got called back into service from reserves to help with the hurricane. So we're trying not to lose them, but I think what Harry's saying is that we tried to give an estimate that could account for possibly having some people who would choose to not start in November but either start in December or January. Jeffrey P. Meuler - Robert W. Baird & Co. Incorporated, Research Division: That's very helpful. And then as we look out to Q1 and the comps ease, would you expect to get back to net course registration from new student growth once you get up against that easier comp?

Yes. We're not giving any guidance for 2013 yet except for the fact that we do think that we will have margin improvement in 2013. And we certainly think these -- the renewal of the Wal-Mart contract and the relationship with New Horizons will prevent growth opportunities in 2013. But other than those general comments, we're not giving guidance right now for the first quarter next year.

Your next question comes from the line of Kelly Flynn of Crédit Suisse. Kelly A. Flynn - Crédit Suisse AG, Research Division: A couple of questions. I think you touched on it briefly in your remarks, but can you just revisit what you're hearing and thinking about the future of military tuition assistance levels and the 90/10 military inclusion issue, in light of the election in particular?

The first one is pretty easy. We've literally heard nothing as far as any changes to TA, and that includes whether or not sequestration would occur if there was some impacts. So if there's something, we actually have a number of different sources that we can access and none of those sources have indicated that anything is up. So that's good. And by the way, one of the sources would be the people who do the billing who have to make changes, and we still haven't heard that they've been given instructions to make changes. As far as the 90/10 goes, there's one bill, I can't remember whose name -- there's like half a dozen bills related to either making schools that participate with VA provide a heck of a lot more data. But there's one bill in particular that's out there that says that they would -- and it's a Senate bill -- that they would include VA and TA in the 90/10 calculation. There's nothing on the House side that I'm aware of because that bill would be dead on arrival in the House Education Committee, thanks to Congressman Kline from Minnesota who chairs it and who happens to be a 20-year Marine Corps veteran. He's totally opposed to it. So that isn't to say though, in a spirit of -- if a spirit of compromise actually surfaces between Chairman Boehner and the President that something couldn't happen. We have certainly done our best, including testimony for Senator Harper's subcommittee -- Carper, I'm sorry. Not Harper, but Senator Carper's subcommittee, where we've said we're opposed to that. And the reason we're opposed to that is because of what most people who cover the for-profit sector know but what most people in Congress don't seem to know, that the…

Sure. It was $871,000. Typically, what we do is we track how we're doing to budget for our students and we saw in the third quarter that our students were coming in a little higher than what we budgeted. We just said there's no need to spend this money. And so marketing deferred $871,000. They're planning on spend it in the fourth quarter. And it's just that we know -- we have this -- our goal was always to keep our marketing expense under 20% for the year. So when we didn't have to spend it, we chose to move it into the fourth quarter. Kelly A. Flynn - Crédit Suisse AG, Research Division: Okay, great. And then lastly, can you just clarify the launch dates of the tech fee and the e-books? I think I had September 1 for the tech fee and Q3 for e-books. Is that still right?

Well, the tech fee we made -- we had an IT decision to make on how to implement it, and it was much easier to build it in since we didn't have a tech fee, that students who registered after a certain date would be charged the tech fee. So it was September 9 or 10, one of those 2 dates. Anybody who registered prior to that didn't get charged the tech fee, and you could register for up to 5 months. Anyone who registered after that date would be charged a tech fee. So that's how that worked. Harry, you want to answer the other part of the question?

Yes, we started to see the improvement from e-books late in the second quarter, and then we're continuing to implement it with additional courses throughout the year. So we're just about on our target of trying to achieve about $4 million of savings this year and $8 million next year. We've gone to -- through the third quarter, we've actually recorded about $2.3 million of savings from the initiative. We've gone from about $65 per class -- per student per textbook per class undergraduate students to about $49 per textbook.

Your next question comes from the line of Gary Bisbee of Barclays.

I guess the first question, you made commentary about you think you've done a good job eliminating the stipend chasers, and certainly, the persistence rate as we calculate it was up nicely in the quarter and would attest to that. But the guidance on starts and total registration seems to imply that per student decline year-over-year or, at best, is flat. I guess, I wondered why. And is there something going on there with the underlying retention rates? Or is it something about the math of how we calculate persistence decline [ph] ?

I think it's probably the math, Gary. I'll let Harry get precise. But the fourth quarter was really the peak of the stipend chasers for a lot of different reasons. We announced that we had an issue with them in the third quarter, but if you'd know, you have to amortize your revenues by day. And so when we go back and look at quarters where we had the greatest impact, even though we were trying to put various procedures into place as early as the beginning of the fourth quarter it was -- our peak was really in the fourth quarter of last year. And our analysis, particularly the way we were able to make an estimate, was to go back and look at the delta or the change in what our typical new student failure rate was versus that particular quarter, and Harry came up with the estimate. You want to talk about the estimate, Harry?

Yes, I -- yes, again, we believe that if you took out the people we think were abuse students last year, our new student guidance for growth would have been 8% to 9% net course registration growth in the fourth quarter this year, and 19% to 20% net course registration growth. So again, I'm not -- and I do the math. I don't see that, that's a reduction on our retention rate, but again it's -- hopefully, the hard part's trying to comp against last year and figure out what the number would have been. All that'll be out of the way by -- when we get into the fourth -- first quarter of next year and as we move throughout next year. So this is a temporary comp.

Yes. And by the way, I think if you don't have that data, which we gave you by making the estimate, I think you might be able to see a change in the retention rate. But that's primarily because of all of these crazy course takers from a year ago.

And so x that, as far as you're seeing, are you seeing the retention improve as those students have dropped and been basically weeded out of your numbers?

Yes. Yes, we are, and we're seeing our pass rates over the first course have improved since last year.

Okay, great. And then next question, can you give us any sense what percent of students have signed up for up to 5 months of courses? Is that a lot of them? And I guess what I'm getting at, would we be better off modeling a fairly modest impact of revenue per student from the IT fee in the fourth quarter but then accelerating as we get into the first half of next year? Or is it likely to have a more noticeable impact, do you think, this next quarter?

Yes, I know -- I think you're right. I think it's going to accelerate next year. It's a little diluted in the fourth quarter this year by the students who preregistered. You won't have that happen next year.

Okay. And then just the last cleanup one. It looked like to me and maybe I just had this wrong from last year, but the bad debt expense comp that you mentioned and put in the 10-Q was actually lower than what you'd reported a year ago. Was there any change in how you're categorizing that? Or did I just have the number wrong? I think you -- I had, had that you said 3.0% in the third quarter '11 was bad debt. Now you're saying 1.7%.

No. No, we're not in the Q.

Yes. I think that should be right out of the Q so...

Your next question comes from the line of Paul Condra of BMO.

I just wanted to ask about the EPS in the third quarter. You had a pretty good upside to your guidance. How much of that is related to the shift in expenses you talked about versus just you coming in ahead of plan?

Well, I think that, that marketing shift was about $0.03 EPS, but the rest of it was just doing a little bit better. Bad debt expense, we didn't expect that to come down quite as much as it did either. So I think just the improvements of the e-books, the decline in the bad debt expense, the 3% timing difference of marketing expense pushing that from the third quarter to the fourth quarter, I think that those were the main differences.

And then I want to follow up on the New Horizons. Do you have any kind of option there in that agreement to make a bigger position if you want to? And then what would persuade you to do so if you do? And just if you could speak a little more broadly about what you want to do with your cash in the future?

Well, I think we've talked about our cash that we know there's a number of options to us if we can't find attractive investments, that we would either continue a stock buyback of some type or perhaps a dividend and that our board wanted to wait until we know more about whether or not there's going to be any changes to the tuition assistance program before we would consider changing either a buyback or a dividend. As far as investments go, Harry and I have talked in the past. We don't have a team to go out and acquire stuff, so nobody should worry that we're gearing up for that sort of thing. It's Harry and me and as things coming over the transom, and we may get the management team involved in looking at an opportunity. In this particular case, we were familiar with New Horizons from having evaluated their training program some years ago for academic credit. And when the leadership there decided that they wanted to find a group of investors who were actually interested in not just selling the company but in growing it going forward, they contacted us, and the more we thought about it, particularly in line with the research we had done about how much it was going to cost us to bootstrap an international marketing effort on our own, we thought it was pretty attractive. So I think we will -- there is no current plan for us to staff up an acquisitions department like other public companies do, and I'm familiar with, and my prior experience in -- as -- in public health care companies. But we'll continue to look at opportunities, and if they should fall in line with our long-term strategy as New Horizons did, then maybe we'll do something. But for the most part, that's not a focus of ours.

Yes. And just to go back to the previous question of -- as I'm looking at the bad debt expense, maybe the confusion is for the 9 -- for -- through September 30 of last year, for the first 9 months of 2011, our bad debt expense was 1.7% of revenue. For the fourth quarter of last year, it was 3.1%. So that was -- I think that was where maybe some of the confusion came.

Your next question comes from the line of Jeff Volshteyn from JP Morgan. Jeffrey Y. Volshteyn - JP Morgan Chase & Co, Research Division: I wanted to go back to the systems and processes that you've put in place for stipend chasers. It sounds like you're still making tweaks to it. Is this -- how comfortable are you with the level of protection and in the balance between overprotecting and the protector at this point? And I can have a follow-up after that.

Well, I think we're pretty comfortable on students who are actually doing this for fraud. I think though -- and I'd love to see some research done by this, but unfortunately, I think there's a number of political interest in not doing the research. But there is a certain percentage out there of students who don't just go to online schools or for-profit schools but go to public schools who know how to take advantage of the FSA process and sign up for classes and get -- do enough work in the class so that they get to the point where the institution grants them the refund that they've asked for. We're never going to get rid of them unless we're allowed to deny dispensing excess funds above what our tuition costs. We had talked last quarter, we've received permission to participate in experimental program that the department's offering where we can actually do a $2,000 reduction in the excess fund request. And that should start, pending some of our technology automation, around July 1 of this coming year -- of 2013, where we can participate in that program. But I think we've got enough checks and balances on the fraud side. Harry, can comment if he thinks I'm wrong. But we do not have an ability to avoid the individuals who cleverly choose to use FSA as a way of paying off credit card debt or borrowing to do other things that they know how to work the system.

Yes, we continue to monitor it. These people are very clever and some -- there are actually some rings of people that are doing this, and we work in conjunction with the OIG. We actually are participating in 3 active OIG investigations to try to prevent these students from abusing the system. And they're clever. They keep changing, as we make changes, they figure out how to get around it. Then, we have to make additional changes, so it's an ongoing process. But we really do think we've stemmed the tide, and that flash mob of abuse students we had come through the second half of last year has gone somewhere else. And we're not seeing that type of situation anymore. Jeffrey Y. Volshteyn - JP Morgan Chase & Co, Research Division: So when you look at the first quarter of 2012 and the second quarter of 2012, what percentage of total starts would you guess? I know it's much lower, but what would it be coming from the stipend chasers as an estimate?

Yes. I don't really have that number right now. It was much reduced by the initiative we started putting in place in January, and it's just a trickle by June. So I'll try to give more color on the future call. I really don't have the numbers in front of me right now. We guessed that rate actually by the difference in the persistent rate of first-time students from our historical persistence rates. So that's how we come up with the estimate. Jeffrey Y. Volshteyn - JP Morgan Chase & Co, Research Division: Okay. And as a follow-up, we hear from conversations with some of the other schools that accrediting agencies are not requiring but suggesting tighter admission policies and sort of more scrutiny of the academic credits of transfer students. Do you see that? And what has been the general tone of conversations with your accrediting agencies?

Accrediting agencies are -- I would say, they have a flexible standard. They don't have a bright standard. But they ask that institutions demonstrate what their admissions policies are and why, so if we're open enrollment as we are for most of our programs -- we do have a couple of programs that are not open enrollment because they're specially accredited programs that require a minimum standard that exceeds our typical standard of graduating from high school or having a GED. But the bottom line is that what they ask is that you're true to your policies and that you can justify that someone that you have admitted is capable of completing college. And I think that's the tricky part, that not all high schools are equal, and most of us who've been totally online have accepted people as long as they have graduated from high school or have a GED. And the fact of the matter is there are people who've graduated from high school that just are not capable of completing college. So with that as a fairly broad standard, our faculty has been working on ways of diagnosing and testing our students in our first required undergraduate class, which we call College 100. And that -- if we make that diagnosis in the first week of class, then we are able to recommend to the student that they be discharged from attending college unless they're willing to pay out of pocket to attend remedial and developmental programs that aren't eligible for Title IV and aren't eligible for military tuition assistance reimbursement.

Yes. And we've -- so we've made that change and actually, the federal government also mandates that students now have to have significant academic engagement that first week of class. And you can tell fairly early on whether the student's capable of doing the work or not. And if they don't complete successfully the significant academic engagement assignments we give them the first week of class, then they're dropped from our system, and we don't count them. There's no harm, no foul. The student doesn't have -- incur any financial penalty. We don't get any financial aid. They're dropped and we don't count them as new students until they've successfully completed the first week. So we kind of have that in place already.

Your next question comes from the line of Adrienne Colby of Deutsche Bank.

I was wondering if you could just comment if there were any timing changes that affected the revenue line either in this quarter and expectations for next quarter just in terms of -- from a calendar perspective?

I don't believe there were any timing differences from a calendar perspective this quarter. As you know, the classes start the first Monday of every month, and we recognize revenue on the days that you attend. So sometimes, if a class starts on the seventh instead of the first, there could be a little difference. I don't believe there were any -- there was any significant impact in the third or fourth quarter.

Great. And I was hoping to just follow up on the comments that you made about the spike you saw in accounts receivable. So it sounded like there were some changes in how the Air Force bases were handling tuition assistance. So I was just hoping you could go back over that and maybe comment if you're expecting any of the other branches to make similar changes.

Well, we're #1 in the Air Force, as you know, and we have a pretty big receivable from them. Prior to June 30, we billed individual Air Force bases every month, and they historically paid within 30 days. During the third quarter, the Air Force changed their payment processing procedures and they centralized their processing into 3 different bases. So now instead billing individual Air Force bases, you have to go through 3 processing centers. When they made that change and we had to change our billings, it ended up in a delay during that implementation period so that our August billings and our September billings were not paid until October. Well, our billings to the Air Force is about $4 million a month, so that was about an $8 million -- over an $8 million increase in our receivable at September 30. And then the Army was -- that -- it seemed to be kind of a more of a one-off thing. We normally bill the Army and collect within 30 days also. But the Army did not pay their $3 million bill for August until the first week of October. So sometimes -- that's a change in the fiscal year. I'm not going to get into the DoDs, why they pay things in 1 year versus another. I don't know. September 30 is the end of their fiscal year. Sometimes, in the past, we've had them speed up payments or delay payments maybe because of their internal -- it really isn't a significant operating problem for us. It's just that if you stop and look at the balance sheet, a snapshot as of September 30, it looks like there's a problem in collection. The other thing I will say is that the Veterans Administration, even though I haven't quantified the number in this release, is also getting a little slower in billing, I think probably because of the volume that they're experiencing. There's been a big increase since the new GI Bill, and the number of people that are taking advantage of veterans benefits. We've seen almost a 100% increase year-over-year in our veterans who are taking classes. So they've been swamped, too, and their billing is a little behind. So it should -- overall, the DoD and the Department of Veterans Affairs billing is a little behind in the fall. I don't think that's -- I think that's pretty typical. That's happened in the past. But I don't think it's a problem. It just looks funny on the balance sheet if you take a snapshot, but it really doesn't impact us.

At this time, we have no further time for questions, and I would like to turn the call back to Mr. Chris Symanoskie. Please proceed.

Great. Thank you, operator. That'll conclude our call for today. We wish to thank all of today's callers for their participation and interest in American Public Education. Thank you, and have a great day.

Thank you for your participation in today's conference. This concludes the presentation. You may now disconnect. Have a wonderful day.