Ladies and gentlemen, thank you for standing by and welcome to the Employers Holdings second quarter 2020 earnings conference call. At this time, all participant lines are in listen-only mode. After the speaker's presentation, there will be a question-and-answer session. To ask a question during the session, you will need to press star then one on your telephone keypad. Please be advised that today's conference may be recorded. If you require further assistance, please press star then zero to reach an operator. I'd now like to hand the conference over to your host today, Ms. Lori Brown, General Counsel. Please go ahead.
Thank you, Liz. Good morning and welcome everyone to the second quarter 2020 earnings call for Employers. Today's call is being recorded and webcast from the investors section of our website, where a replay will be available following the call. Presenting today on the call will be Douglas Dirks, our Chief Executive Officer, Mike Paquette, our Chief Financial Officer, and Steve Festa, our Chief Operating Officer. Statements made during this conference call that are not based on historical facts are considered forward-looking statements. These statements are made in reliance on the safe harbor provision of the Private Securities Litigation Reform Act of 1995. Although we believe the expectations expressed in our forward-looking statements are reasonable, risks and uncertainties could cause actual results to be materially different from our expectations, including the risks set forth in our filings with the Securities and Exchange Commission.
All remarks made during the call are current only at the time of the call and will not be updated to reflect subsequent developments. The company also uses its website as a means of disclosing material non-public information and for complying with disclosure obligation under SEC's Regulation FD. Such disclosures will be included in the investors section of the company's website. Accordingly, investors should monitor that portion of the company's website in addition to following the company's press releases, SEC filings, public conference calls, and webcasts. In our earnings press release and in our remarks or responses to questions, we may use non-GAAP financial metrics. Reconciliations of these non-GAAP metrics to our GAAP results are included in our financial supplement as an attachment to our earnings press release, our investor presentation, and any other materials available in the investors section of our website.
Now I will turn the call over to Doug.
Thank you, Lori, and thank you all for joining us today. Our second quarter results were highly favorable, and our accident year results were broadly in line with our expectations. Although we expected to see our new business writings down sharply in the quarter because of lower submission volume, we were pleased to see a rebound in June when compared to April and May levels. We are hopeful that this is a sign that businesses in our targeted classes are beginning to reopen and are resuming their operations, albeit with lower payrolls than they had previously. We also believe that the strategic investments we have made over the last several years in data analytics and technology have in fact enabled a compellingly superior ease of doing business, further contributing to our growth.
Our results at an individual state level reflect the significant differences in the depth and breadth of post-shutdown reopenings. We see this particularly in some of our hospitality classes, where some states have returned to normal while others remain almost completely shuttered. In some states, we are experiencing business activity that is on par with or exceeds what we observed immediately prior to the pandemic shutdowns. While in others, most notably and significantly California, activity levels continue to lag. Excluding California, our policies in force were up 5.5% over the previous quarter and up 11.3% year to date, while in-force premium was down a mere eight-tenths of a % for the quarter and down 1.3% year to date. We have been impacted by regulatory actions that have either mandated or requested that we suspend cancellation of policies for non-payment of premium.
These orders or requests were for differing lengths of time, varying by jurisdiction, and now have mostly expired, permitting us to resume routine cancellation activities. Our year-to-date results reflect actual and anticipated increases in uncollectible premium and bad debt, although it is still too early to estimate the ultimate cost resulting from these orders. We expect that both actual policies in force and in-force premium will be lower than our current count because of these moratorium and the opaqueness they have created. We will have greater clarity on their impact in the coming weeks.
Unlike most other lines of property and casualty insurance, where pandemic-related changes in exposure resulted in broadly applied premium credits, workers' compensation is self-adjusting to actual exposure for the policy period, either through midterm endorsements or final audit adjustments. Midterm premium endorsements processed in the quarter were a positive $1.1 million, a reduction of $5.6 million year over year. Final audit adjustments were a negative $4.4 million, a reduction of $11.1 million year over year. Combined for the quarter, these two exposure-adjusting components reduced written premium by 9.5% year over year. Unlike most other lines of insurance, workers' compensation benefits are defined by statute, and consequently, cannot be changed by us through policy terms, but rather, can only be changed through legislative action or judicial interpretation. In many states, insurance commissioners, legislatures, and governors have retroactively expanded definitions of compensability and created new presumptions related to virus exposure.
Many of these changes have been limited to first responders and frontline healthcare providers. Some states, however, have adopted more expansive categories of workers entitled to compensability presumptions related to COVID-19 exposures. These changes will have a negative impact on ultimate losses for the workers' compensation industry, although we continue to believe our exposure to additional losses from currently enacted changes are likely immaterial given the classes of business we write. It is important to note, however, that the materiality of these changes could be greater if the presumptions are further expanded or the time periods during which they apply are lengthened. In the quarter, we recorded $24 million of favorable prior year loss reserve development, which related to nearly every accident year. The favorable change in reserves that we recognized this quarter related solely to development that has emerged since March 31st, 2020.
You may recall that last quarter, despite observing favorable loss development in nearly every year, we recognized observed reserve redundancies only for years 2010 and prior, as we believe those years have relatively low exposure to negative recessionary impacts. As a result, our current reserving position continues to reflect our view that there's a higher degree of uncertainty in the loss reserves of more recent years as a result of a recession. We have invested significantly over the last several years in an operating model that drives superior customer experiences and enhanced efficiencies. As our agents and insureds have adjusted to a different and more challenging operating environment, we believe the solutions we provided them are resulting in more business opportunities for us and more durable relationships with our partners.
We've been fully functional since we closed all of our buildings to employees and the general public in March, continue to operate in a work-from-home mode in order to protect the safety and wellbeing of our employees, their families, and our stakeholders. We feel that Employers is in a strong position to weather the challenges created by the pandemic and to potentially benefit from a different set of opportunities. Our strong financial position and operational capabilities allow us to continue providing superior service to our agents and insureds without disruption. With that, Mike will now provide a further discussion of our financial results. Steve will then discuss some of the current trends, then I'll return for a few brief closing remarks. Mike?
Thank you, Doug. During the second quarter, we delivered an 8.9% annualized return on adjusted equity, which is a terrific result under the current circumstances. Our underwriting results were solid for the quarter, Our top line was adversely impacted by a decrease in new business premium, a reduction in estimated final audit premium, additional, but moderating premium-reducing midterm endorsements. Our net premiums earned were $152 million, a decrease of 14% year-over-year. Premiums earned are primarily a function of the amount and timing of net written premiums, I'll let Steve describe the decrease in premium writings that we have experienced in his remarks. Our loss and loss adjustment expenses were $73 million, a decrease of 16%, which was primarily due to the decrease in earned premiums.
As Doug previously mentioned, we recognized $24 million of favorable prior year loss reserve development during the current period, which was recognized across nearly every prior accident year. Our current accident year loss and LAE ratio was 65.5%, which is unchanged from our first quarter and full year 2019 indications. Commission expenses were $19 million, a decrease of 19% year-over-year. That decrease was primarily due to the decrease in earned premiums. Underwriting and general administrative expenses were $45 million for the quarter, consistent with that of a year ago. From a reporting segment perspective, our Employers segment had underwriting income of $18 million for the quarter versus $23 million a year ago, and its combined ratios were 88% and 87%, respectively. Our Cerity segment had an underwriting loss of $4 million for the quarter, consistent with its underwriting loss of a year ago.
Turning to investments, our net investment income was $20 million for the quarter, down 7%. The decrease was primarily due to a sharp decrease in the amortization of bond premiums associated with our residential mortgage-backed securities, which was caused by an acceleration in mortgage loan prepayment speed assumptions. At quarter end, our fixed maturities had a duration of 2.7 and an average credit quality of A-plus, and our equity securities and other investments represented 9% of the total investment portfolio. We were favorably impacted by $69 million of after-tax unrealized gains from fixed maturity securities, which are reflected on our balance sheet, and $25 million of net after-tax unrealized gains from equity securities and other investments, which are reflected on our income statement. These net unrealized investment gains were the primary driver of our more than 11% increase in book value per share, including the deferred gain during the quarter.
Finally, during the quarter, we repurchased $31 million of our common stock at an average price of $29.76 per share. We repurchased a further $3 million of our common stock thus far in July at an average price per share of $29.21. Our remaining share repurchase authority, which was increased by $50 million earlier this week, currently stands at $52 million. Although our third quarter repurchase activities will be tempered as compared to those of the first two quarters due to the timing of planned subsidiary dividends to the parent. Now I'll turn the call over to Steve.
Thank you, Mike, and good morning. Net written premiums for the quarter of $139 million were down $36 million, or 20.7%, from the second quarter of 2019. The primary drivers for this decrease were new business premium and final audit premium. New business bound policies were down for the quarter relative to the second quarter of 2019, driven by decreases in the months of April and May. This was driven by lower submission volume for these two months as we saw less accounts going to market, which is what we expected due to the COVID-19 impact. This trend changed in June, which reverted back to strong year-over-year growth in both submissions and bound policies. For June, new business submissions were up year-over-year by 4.8%, and new business policies bound were up 14%. However, we saw a decrease in average premium size on new business bound for the quarter.
This decrease was greater than seen in previous quarters, and we attribute this larger decrease primarily to lower than usual payroll due to COVID-related impacts to business. With respect to renewal business for the quarter, we continue to see high policy unit retention rates. For the quarter, the rate was 95.3%, an increase from 93.9% in the first quarter. This was offset to some degree by continued rate decreases. Renewal premium was down 3.8% when compared to the second quarter of 2019. Total written premium endorsements decreased quarter-over-quarter by $5.6 million, driven primarily by endorsements related to payroll reductions. The peak of those payroll reductions was seen in April, with May and June showing a slowing down of those reductions. Final audit premium was reduced quarter-over-quarter due to decreases in actual and projected payroll. New claim volume continues to decline on a year-over-year basis.
In comparing each month of the quarter to the prior year, April exhibited a 49% decrease in lost time claims, May decreased 41%, and June was down 11%. These decreases are driven by less exposure caused by businesses being shuttered as well as lower headcount and hours worked in businesses that remained open. Now I will turn the call back over to Doug.
Thank you, Steve. As was the case last quarter, we continue to believe that the COVID-19 pandemic is more likely to be a premium event than either a capital or a claims event for workers' compensation. We continue to carefully monitor political intervention in expanded definitions of compensability and to actively engage in public policy discussions where appropriate. Although we are experiencing less volatility today than we did in the first half of the second quarter, we expect that new business production and renewal premium will continue to be somewhat uneven as states either allow businesses to reopen or reinstate business closures. With that, operator, we'll now turn the call over for questions.
As a reminder, ladies and gentlemen, if you'd like to ask a question at this time, please press the star, then the number 1 key on your touchtone telephone. To withdraw your question, press the pound key. Our first question comes from Matthew Carletti with JMP Securities. Your line is now open.
Hey, thanks. Good morning. I have a couple questions. Hoping I could start off, Doug, maybe on pricing and loss costs. I know it's hard to strip out everything going on with COVID-19, but how do you feel about the market here in terms of the frequency and severity trends, maybe adjusted for premium or payroll falloffs, and where pricing is, and maybe if you could separate California, outside of California. One of your big competitors, I think, called the bottom of the market yesterday. I hope that's the case. Just curious where you guys are seeing it.
Let's start with the observation that going into this year, virtually every state in which we do business had a filed rate decrease. There's nothing in any of what's happened over the last six months that changes that. Just in the ordinary course, we would've expected to see continuing headwinds on average rate. In fact, there's no reason to think that has changed. To the extent that anyone is observing any strengthening in pricing in the market, it would be more a reflection of the competitive environment than it would have to do with the underlying loss costs and rates. It continues to be, for us, a very competitive environment. As we've observed in the past, that's particularly true on larger accounts. We continue to see every bit as much pricing competition today as we were seeing three and six months ago.
At least from our perspective, the market hasn't yet changed. Why don't I turn this over to Steve? I'll let Steve talk a little bit about what he's seeing in terms of loss trends.
As I mentioned earlier, Matt, what we're seeing is the decrease in lost time claim incurs. We saw that happening even before COVID-19. The first two months of this year, we saw a decrease in our incoming claim volume. That's obviously continued. It's accelerated in this quarter. It's too early to call out what's happening in July, but clearly, with the reduced exposure that we have with less employees working, we expected to see this. What we don't have a handle on, and won't for some time is, as this recession continues, will we see a replication of what we saw in the last recession with respect to post-termination CT claims? In particular, the industry saw that quite a bit in California. That's something that we're going to be obviously monitoring going forward.
From a severity standpoint, we're not seeing anything with respect to changes in average severity. We're not expecting to see that change in the near term either.
Got you. My one other question is, just as we think about the top line, obviously, there's a lot of forces at work. I appreciate, it's really nice that you provide the monthly numbers in terms of seeing the new business production coming back. Is there any way you can help us think about, obviously some states, California notably, had moratoriums on cancellations and things like that. How should we think about that impact as we progress into Q3? Do you think Q2 has seen the worst of it, or there's some catching up to do? I don't know if you can help us think about maybe how much of the book in Q2 maybe wasn't current on payment, if that's been a big change or a small change from what it would normally run at.
Any way to help think about that piece of the pie would be helpful. Thanks.
Matt, you may recall that we at Employers made a decision to put a cancellation moratorium in place above and beyond what some states required. That moratorium was lifted. It was put in place early on in the COVID-19 crisis, that was lifted on June 15th. Post June 15th, we sent out notice of cancellation letters to those insureds that had been in a holding pattern from being canceled because of our moratorium. The states that we do business in have different requirements with respect to how long an insured has to make a payment upon receipt of the notice of cancellation letter, and that varies by state. We'll have a better handle on how much of an impact that lifting of the moratorium will have probably in the next two to three weeks.
It's early enough now that we don't have a good handle on that. There are a small handful of states that have moratoriums in place that extend beyond what ours was, but that's a small percentage of our overall premium. We'll have a better handle at the next two to three weeks on what that lifting of the moratorium will do to our in-force business.
Okay. All right. Does the business you have with ADP or that direct reporting, does that help in this exercise where you can see some of that in real time come through the payrolls? Are you able to use any of that kind of data that you see on that part of the book to try to extrapolate it to the rest of the book in Q2? Do you really just have to wait to see what happens in Q3?
Yeah. The benefit we get from the 25% of our business that's driven through pay-as-you-go programs, ADP and Paychex being the largest, is that we get some "real time" within two weeks, visibility into what's happening with payroll. What we don't have at this point is a better understanding, which we will in the future, of how many businesses are lowering their payroll because they've reduced staff temporarily versus those that have reduced payroll on a permanent basis because they will be closed. That's the data point that we'll have some perspective on in the near future that we don't have today.
Okay, great. Very helpful. Thank you for the answers and best of luck going forward.
Thank you.
Our next question comes from Mark Hughes with SunTrust. Your line is now open.
Yeah, thank you. Good morning. Just following up on that. It seems like we're several months into this. I just wonder how much variability there could be when gauging that in-force premium. Presumably, you would've heard from most folks who are under distress. You're getting a lot of inputs that seem to point to stabilization. How much variation could there be when we think about the in-force?
I think some of the uncertainty, Mark, is we saw what happened in late March, mostly April, really peaking in April. As states start to allow reopening, we're seeing some of that impact as well, as business returns to something more like it was at the beginning of the year. I think the challenge is with states opening and closing on very irregular schedules. Depending on the state you're in, it's very difficult to project the impact. One of the things that's really interesting is that California continues to be very challenging. Obviously, California has been in a fairly strict shutdown mode for some time now, although I'm hearing that there's much more activity in California today.
At the same time, although Florida's being hit very hard right now with new positive tests, our business flow in Florida is quite strong and on a day-to-day basis, doesn't seem to be being impacted by what we're hearing about in Florida. It is extremely difficult to project what activities are causing what type of either acceleration or slowdown in business activity. I will call out again, notably, California is the most challenging environment we're seeing anywhere in the country. It's been tough in the Northeast, but it appears to be getting better now. The balance of the states are flat to growing for us. Absent a rather severe shutdown of the entire national economy again, we're somewhat optimistic that we've seen the worst of this. To project it is extremely difficult.
Under those circumstances, is it fair to think that it wouldn't be maybe more than a few points if you do see some variability on that in-force?
I think the challenge with the in-force is, I referenced it in my comments about the opaqueness, to the extent that we have had a moratoria across the states on cancellation for non-payment activities. In some cases, we don't know if that business is even in business anymore. If they are in business, if they're going to come back with a significantly lower payroll. Maybe at that point in time, they'll endorse back down to what their expected payroll is. It is exceedingly difficult to get in the mind of 104,000 policyholders and know which ones are still in business and which ones aren't. You probably saw the article that came out referencing what Yelp is observing, and we pay very close attention to all of those. That would suggest that a lot of businesses in the classes we write are gone.
We're paying attention to that. We're simply not seeing it, part of it because of the way workers' compensation is structured and the way states have extended periods of time for policyholders to go without paying. I do expect, however, that we are over-counting in our current count, the number of policies that are in force and what in-force premium is, simply because I can't imagine those numbers will go up. They can only go down. How much they go down is fairly uncertain. I'll go back to the question related to ADP and our payroll partners, Paychex, the others. We do keep an eye to that to try to understand what type of activity those policies are exhibiting in real time. Trying to apply that against the other 75% of our business that is being written on a guaranteed cost.
When we were thinking about what the adjustments might be to things like premium audit, bad debt, and the like, we did take into consideration what we were seeing in other parts of the business. It's not that we don't know what it is and we're consequently ignoring it. We actually have already taken it into consideration, albeit not directly.
I'm sorry, did you say what proportion of your premium base is subject to the moratorium, or you haven't had the premium payment?
Mark, yeah, we self-imposed our own moratorium that ended on June 15th. There are about four states left that have a cancel moratorium that extends beyond June 15th, but I don't have the % of the premium, but they're states that make up a very small % of our in-force premium.
I'm sorry, I was thinking of your policy holder base, what proportion has not paid during this moratorium period?
It's a single-digit number. I don't have that in front of me. It's a relatively small percentage in terms of the number of insureds. I don't have that exact number in front of me right now.
Yeah. I kind of get back to it seems like a few points. I won't belabor the issue. When you think about the comparisons in the coming quarters in terms of the audit premium or midterm endorsements, assuming you have your flat on both categories in the second half, how much of a headwind is that from a year-over-year perspective, or how much of a benefit was it in last year's second half?
Certainly, the headwind continues. Mark, that's a difficult question for us to answer because it's going to be influenced so much by when the recovery truly begins, how rapid it is. What I suspect is happening, and Steve referenced it in respect to new business, even though we're seeing more units and more binds, the average premium is down. I suspect that's because those policies are now being essentially priced in real time. I would expect that most businesses at renewal are providing a more pessimistic view of what their likely payroll is going to be for the next 12 months. The reason I call that out is because at some point in time, audit premium probably swings back around to being a positive number because the economy is getting better, payrolls are increasing, and the initial estimates were low.
That will be a function of when the economy recovers and how quickly payrolls come back. In some instances, we do see insureds come in and endorse their existing policies to add classes or increase payroll because they want to remain current. In most cases, we pick that up at final audit, and the final audit cycle is 12 to 15 months out post policy inception. If you think about it, business we were writing in April, we're not going to be seeing the audit results until a year from now, and it's just extremely difficult to project that number.
I definitely sympathize on that. I'm thinking, what was the year ago actual when we think about Q3 and Q4, if we're just thinking about trying to do, what's your in-force premium, what are going to be the pluses and minuses in coming quarters, but then trying to think about that in connection to last year's second half. I was just sort of curious if you have the specifics on last year's second half, what those adjustments equated to.
We don't have that immediately in front of us, Mark. That's something we'd be happy to follow up with. I think to the comments that we made earlier, we are seeing endorsements, premium-reducing endorsements fall over where they were early in the second quarter. If things continue on the track they're on, I don't see any reason to think that's going to go up. It's not yet returned to what we would've described as normal. If you'll look at the June numbers, certainly much better than April and May, but still worse than what we would've seen in January and February. They've not quite returned to what we would call normal yet, although they're just a fraction of what they were in April. In terms of premium audit, that number is always a challenging number for us.
As I indicated, I think we'll probably have maybe a quarter or two of pressure on premium audit. For those policies that haven't endorsed their payroll down, but have in fact decreased their payrolls, we'll be seeing that as they come up for audit now. I think when we start getting out into the beginning of next year, we would expect that if in fact those payrolls were pessimistically stated, we'll start seeing lift again from premium audit.
A final question. Steve, did you mention what the new business trends have been like in July so far?
I didn't mention it, Mark, but I can tell you that we're seeing, very similar to June, an increase in submissions and an increase in bound. In fact, in terms of at the halfway point through the month, the increase in bounds was at a higher % than we even saw in June. A positive trend replicating what we saw in June.
Thank you very much.
Thank you.
As a reminder, ladies and gentlemen, that is star then one if you'd like to ask a question at this time. Our next question comes from the line of Bob Farnham with Boenning & Scattergood. Your line is now open.
Yep, thanks. A question on California. You're talking about it being pretty competitive or the most competitive state. Are you seeing any other of your peers raising rates there yet? Or is that kind of one of the primary reasons why you're having trouble getting new business there, is just the rates are higher?
No, Bob, we have not seen that happen. In fact, the market is as competitive as it ever has been. In fact, I would tell you that over the past few months, we've probably seen even heightened competition from a pricing standpoint on the larger accounts, definitely the middle-market business. We are not seeing a turn in the market in California from a competitive pricing standpoint.
Okay. One other question I have on the accident year loss ratio. You're talking about pretty steep declines in claims frequency, at least in April and May. Still, kind of kept that accident year loss ratio level at 65 and a half. I'm just curious if you're thinking that's going to end up being conservative over time or what your thoughts are on that.
Bob, what I'd say there is, I think it's too early to tell. Even though we've got some data that probably looks like that number would be favorable, presumption can change, COVID cases can change, we're keeping it at the level that we're comfortable with for right now, despite the fact that kind of the actual claims levels that we're seeing right now could argue for a decrease in future periods.
Yeah, I guess when theoretically states can retroactively change things, makes you a little more conservative with it. Okay, that's it for me.
I'm showing no further questions in queue at this time. I'd like to turn the call back to Douglas Dirks for closing remarks.
Thank you, operator. Thank you everyone for joining us today. We were quite pleased with the performance in the quarter. Although things are challenging, we're fairly optimistic about how we're positioned going into a recovery. We hope you're all well and stay safe. We look forward to talking to you next quarter. Thank you all very much.
Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect.