Good day, everyone, and welcome to AMERISAFE's 2019 Fourth Quarter and Year-End Earnings Conference Call. Today's conference is being recorded. At this time, I would like to turn the conference to Kathryn Shirley, Chief Administrative Officer. Please go ahead, ma'am..

Good morning. Welcome to the AMERISAFE 2019 Fourth Quarter Investor Call. If you have not received the earnings release, it is available on our web at www.amerisafe.com. This call is being recorded. A replay of today's call will be available. Details on how to access the replay are in the earnings release.

During this call, we will be making forward-looking statements. These statements are based on current expectations and assumptions that are subject to various risks and uncertainties.

Actual results may differ materially from the results expressed or implied in these statements if the underlying assumptions prove to be incorrect or as a result of risks, uncertainties and other factors, including factors discussed in today's earnings release, in the comments made during this call and in the Risk Factors section of our Form 10-K, Form 10-Qs and other reports and filings with the Securities and Exchange Commission.

We do not undertake any duty to update any forward-looking statement. I will now turn the call over to Janelle Frost, AMERISAFE's President and CEO..

Thank you, Kathryn, and good morning, everyone. The current soft market is unique and its challenges, given the state of the property and casualty industry. Over the past year, there's been a lot of focus on the continued rate declines in workers' compensation. And yes, the impact is significant to the industry.

Carriers are collecting less premium for a growing exposure base. However, there has also been favorable trends. Payrolls expanded with a strong economy, frequency trends were stable and medical inflation tempered.

These favorable trends, coupled with our distinctive approach to workers' compensation, produced another year of excellent financial results for AMERISAFE. We reported a combined ratio of 76.6% and a return on average equity of 22.1% in 2019, the fourth year of workers' compensation rate declines.

For AMERISAFE, less than anticipated average loss severity for prior years offset the impact of lower premiums. Our gross premiums written in the quarter were down 6.1% from the fourth quarter of 2018. Strong payrolls and audit premiums lessened the impact of rate declines and a slight drop in overall policy count.

Our retention for policies for which we chose to offer renewal remained strong at 94.1%. However, new business finds were not enough to cover lost renewals in terms of policy count. Our pricing for the quarter, as shown in our ELCM, was at 159 compared to 162 in the fourth quarter of 2018.

Back to those trends I opened the call discussing and the impact to losses. First is frequency. I choose to talk about frequency in terms of earned premium. For AMERISAFE, we anticipated flattening to slight increases in frequency for the last few accident years, more a function of decreased premium than actual claim counts.

For example, our reported claims for accident year 2019 were down 4.4% from the accident year 2018 at 12 months. While earned premium for the calendar year was down 5%. I still believe that in a full-employment economy, we have more unskilled workers in high-risk industries, which will lead to an increase in workplace injuries.

For these reasons, I anticipate there to be continued pressure on frequency. The second loss trend is severity. Average severity was distinctly impactful to this quarter's results as it relates to prior accident years. I believe we've been transparent of how we think about the more recent accident years, often referring to them as green.

Our consistent book of business provides us rich data based on historical patterns. We rely heavily on our case reserves, which were established timely and with expertise gained through our experience in specialization.

Given the types of injuries we deal with, 30 to 36 months of maturity gives a better indication of how an accident year will ultimately develop. This quarter, favorable case reserve patterns led to an -- recognition of lower average severities than previously anticipated for accident years 2014 to 2017.

In the quarter, the financial impact was a reduction of loss and loss adjustment expense of $26.1 million or 31.7 loss ratio points. As for the current accident year, the loss ratio was 72.5%, and it remained unchanged throughout 2019. We did not adjust our assumptions for those accident years, '18 or '19, due to their greenness.

We also do not anticipate a change in the loss ratio for accident year 2020, barring some material change between now and the end of the first quarter. I'll now turn the call over to Neal to discuss the financials..

Thank you, Janelle, and good morning, everyone. For the fourth quarter of 2019, AMERISAFE reported net income of $34 million or $1.76 per diluted share compared with $18.8 million or $0.98 per diluted share in last year's fourth quarter.

Operating net income was $32.8 million for the quarter or $1.70 per share, an increase from $1.07 in the fourth quarter of last year. For the full year 2019, AMERISAFE produced net income of $92.7 million or $4.80 per share. Operating net income for the full year 2019 was $89 million or $4.60 per share.

This level of operating income is our highest year ever. And was influenced by benign severity trends, which drove significant prior year favorable reserve development as Janelle mentioned earlier. Revenues in the quarter were down 2.9% to $91.9 million compared with the fourth quarter of 2018.

Net premiums earned decreased 7.4% to $82.3 million when compared to last year's fourth quarter. For the full year, net premiums earned were off 5%, totaling some $332.9 million. These trends were driven by the continued decline in workers' compensation loss costs in 2018 and 2019. Turning to net investment income.

We saw a slight decrease of 0.3% in the fourth quarter to $8 million compared with $8.1 million in the fourth quarter of 2018. Net investment income for the full year was up 6.7% to $32.5 million compared with $30.5 million in 2018 due to slightly higher average investment yields.

The tax-equivalent yield on our investment portfolio was 3.12% at year-end and almost unchanged from 3.15% at year-end 2018. The pretax yield on the portfolio at year-end was 2.79%. There were no impairments on any of the securities held in the portfolio during the quarter or for the full year of 2019.

And additionally, there were no significant realized gains or losses during the quarter or the full year. The investment portfolio is high quality, carrying an average AA rating with current duration of 3.95. And the portfolio is comprised of 61% in municipal bonds, which includes taxable municipal bonds; 21% in corporate bonds; 12% in U.S.

treasuries and agencies; 2% in equities and the remainder in cash and other investments. Approximately 58% of our bond portfolio is comprised of held-to-maturity securities, which were in an overall unrealized gain position of $21.9 million at year-end.

These gains are not reflected in our year-end book value as these bonds are carried at amortized cost. Moving now to operating expenses. Our total underwriting and other expenses were $14.7 million in the quarter compared with $19 million in the fourth quarter of 2018.

The decrease in operating expenses was primarily due to a $3.5 million benefit as the result of an end of a multiple injury fund assessment. By category, the 2019 fourth quarter expenses included $7.9 million of salaries and benefits, $6.2 million of commissions and $500,000 of underwriting and other costs.

Our expense ratio for the quarter was 17.8% compared with 21.4% for the fourth quarter of 2018. For the full year 2019, operating expenses decreased $6.8 million or 8.4%, heavily influenced by a reduction in insurance assessments, including the fourth quarter benefit mentioned earlier.

As a result, our expense ratio for the full year was unusually low at 22.3% compared with 23.2% in 2018. Our tax rate for the fourth quarter was 20.8% and 19.8% for the full year, both higher than in 2018 as a result of stronger underwriting profitability.

Return on average equity for the fourth quarter of 2019 was 30.3% compared to 17.3% for the fourth quarter of 2018. For the full year, ROE was 22.1% compared with 17.2% last year. Operating ROE for the quarter was 30%. Operating ROE for the full year was 21.5% compared with 17.9% in 2018. And now to capital management.

During the fourth quarter, the company paid its regular quarterly cash dividend of $0.25 per share as well as an extraordinary dividend of $3.50 per share. This quarter, the Board of Directors declared a quarterly cash dividend of $0.27 per share, payable on March 27, 2020 to shareholders of record as of March 13, 2020.

This represents an 8% increase in the regular quarterly dividend. And finally, just a couple of other noteworthy items. Book value per share at December 31, 2019, was $22.29, up slightly compared with last year's $21.26 per share. And we paid out $4.50 per share in dividends to shareholders throughout the year.

Our statutory surplus was $360 million at December 31, 2019 compared with $384 million at the end of 2018. And finally, we will be filing our Form 10-K with the SEC next week. That concludes my remarks. And we would now like to open the call up for the question-and-answer session.

Operator?.

[Operator Instructions]. We'll go first to Randy Binner with B. Riley FBR..

So I think we'd like to kind of dig into the expenses to better understand how that would run rate into the future. So just starting with the second injury fund. I think that was quantified at $3.5 million.

Is that -- was that something that was coming out in the fourth quarter? Or was that kind of ratable throughout the year?.

No. That was in the fourth quarter. It was a reversal of an accrual that we had up for the multiple injury trust fund. But the trust fund that is being shut down or has shut down, did its actuarial analysis in 2019 and indicated that there would be no further assessments. So we had $3.5 million up for those future assessments.

And we took that down during the quarter..

So that's a one-timer.

And then that -- I'm sorry, which fund was it? What state was it for?.

It was for South Carolina..

And that fund is no longer -- it's shut down effectively.

Is that right?.

That is correct. They [indiscernible] future assessments. So that's why we took the full accrual down..

Okay.

So that had been -- but that hadn't been -- that had been accrued for, I guess, over time through expenses?.

Correct..

Okay. Then the other question is just -- so if we add that back, then that would be more of the normal run rate. But it still seems a little -- like you were a little lower quarter-over-quarter, maybe on just general compensation expense.

Was there anything there that was lower? Or is there something else that's affecting the year-over-year comparison ex the second injury fund reversal?.

No. Typically, you'll see our fourth quarter expenses are a little bit lower as we accrue assessments throughout the year for loss-based assessments and other types of guaranteed fund assessments. And then in the fourth quarter, if those assessments have not been charged by the states, we will reverse those.

That's why our fourth quarter expense ratio is typically a little bit lower. I would say, generally, in the past two years, we've seen some favorable trends in those assessments being lower than expected.

And I think that's because the trends -- the favorable trends in the workers' comp market, in general, have impacted those assessments because of the favorable severity trends and other trends that we see..

Okay. And then I just had one on medical loss inflation. Obviously, the reserve result was benign for the '14 to '17 accident years. And I think it usually takes about -- like 36 months, 24 to 36 months to season the workers' comp book.

So give me a view, though, on how medical loss inflation and other severity trends are trending now, like in 2019 and 2020, just as you observe the business?.

Yes. I think when we established the 72.5% for 2019, in that assumption, we had mid-single-digit medical inflation expected for '19..

Okay.

And that would be similar to 4Q '17 [indiscernible] net income or?.

Yes. So that would be similar to what we had accrued in '18 and '17..

Our next question will come from Matt Carletti with JMP..

A few questions. One, hopefully we could start with prior period development. I know you mentioned '14 to '17 were the years that it came from.

Can you be just a little more specific across those 4 years? Was there any year or grouping of years that drove the bulk of it? Or was it fairly even across the 4 years?.

Yes. I'll give you the numbers by accident year for the quarter. So accident year '17 was $6.3 million. '16 was $7.5 million. '15 was $6.2 million. '14 was $2.3 million. And then the remainder of the balance was prior to '14..

Got you. Okay, really helpful. Great. And then just shifting to your comment about expecting a flat accident year for 2020, absent some macro surprise.

Am I thinking about that right, that your prior methodology of -- on the frequency view of expected frequency to kind of increase on your earned premium view still holds? And that the offset to that is a different view of severity as maybe informed by the prior predevelopment that's been taking place?.

Yes. I think that's a good way to look at it. I mean, as I said in my prepared remarks, we -- I still believe that we're going to see an uptick in claims just as unskilled workers are working in the hazardous industries. So that should drive frequency -- at least flatten to slight increases in frequency. That hasn't changed.

But we have seen, as I mentioned, in the severity front, we have seen severities be less than we anticipated, again, the accident years '14 through '17. We're not willing to call it, as I like to say, on '17 and '18 just yet. But I see -- or yes, '18 and '19. But I think looking into 2020, we expect that to be relatively flat with '19.

You've followed us for a long time, Matt. I think we're pretty -- we're quicker about recognizing things when we see negative trends or negative things in the data, hearken back to accident year 2017. During the accident year, we raised our loss pick as we saw an uptick in frequency at that point.

I think we're a little slower to -- or a little bit more cautious about recognizing more positive trends in our data. This is a long tail line of business. It's lumpy. So we don't try not to be too ready -- quick on the draw when it comes to the positive side of that..

Yes. Definitely served you well over the years. Last one, if I can, just on capital management, more just on methodology and then strategy.

Am I right in thinking that assuming we're still in a flat to down revenue environment because seems like there's still pressures there and absent, say, a turn in the market, that's probably the case in the foreseeable future, that it would be expecting kind of earnings plus to be returned in some form of dividend? Is still kind of the norm that -- to keep the capital ratios kind of stable or improving your absent growth?.

Yes. Matt, this is Neal. We don't typically comment on future dividends. I would point out, we did pay out $4.50 in dividends this year. And we ended up having operating earnings of $4.60. So we sort of outearned the dividend pace.

But management and the board addresses the company's capital position and looks at it each quarter, and periodically make adjustments, which in recent years, have included special dividends..

[Operator Instructions]. We'll go next to Mark Hughes with SunTrust. .

The number of large losses for 2019.

Do you have that number and the comparison with '18?.

Sure. We ended the year with 16, what we would call severe claims, claims over $1 million of case incurred. That compares to 18 at the end of the calendar year for accident year '18, and 17 for accident year '17..

And then how do we think about the IBNR? I think you've emphasized that your savings here and the better performance has been on case reserves.

How much is sitting there in IBNR? And when do you evaluate that?.

Yes. So our historical pattern has been -- obviously, when we begin an accident year, we have an anticipated loss ratio. We tried not to -- unless there's something compelling in the data, that doesn't really change during the current accident year. Although, as I mentioned earlier, harking back to '17, we did change it.

But at that point, we try and list just something completely compelling in the data or really a handful of really large losses that move the needle. We typically don't adjust our IBNR balances until we get to that 30- to 36-month window.

Hence, the changes that you -- we saw this quarter with the favorable case development that we've been experiencing throughout the year in accident years, '14, '15, '16.

You saw some adjustment in the -- as I mentioned in my remarks, the anticipated severities were lower than -- or the actual severities were coming in lower than we anticipated, hence we made the adjustment..

Right.

To case reserves?.

And IBNR..

And IBNR. Okay. The loss cost multiplier for 2020.

When you look at your footprint, kind of roughly what is the expectation on loss cost trends? And this would be an official loss cost?.

Yes. Yes, understood. So our rate in -- the rate that we charged our clients if -- and we talk about ELCM a lot, the premium per 100, which is something we disclosed in our 10-K. The decrease in '19 was around 9.1%, which oddly enough, is about what, I think NCCI projected for the loss cost changes around that number for 2019.

So we were, "In line with that." Their expectation for 2020, I believe is -- I don't think they've published that as of yet. But I would expect it to be higher single digit, based on just the rate filings we've seen thus far..

Right. When I think about your loss pick outlook to hold steady, I guess, on the one hand, you've got competition in the market, you've got declining loss costs, your ELCM numbers have continued to come down slowly but would still come down. All that would suggest pressure on current accident year.

But it sounds like since the severity, and I would assume frequency has been more benign than expected the last couple of years, your people to kind of restrain that upward pressure, and therefore, hold it steady.

Is that the right way to think about it?.

I think that's a very accurate way of summarizing it, yes..

Right, which is to say, at least so far, so good on '18 and '19?.

Yes. Really, that's how I should have said it. Yes, they're still very green for us, and we're not willing to -- so, yes, this is all good, but you're right. So far, so good..

Yes. Yes. And then the audit premium was up a bit.

Any general comment on payroll? I don't know whether you've made any comments about that in your earlier remarks, but the trends in wages audit?.

Right. We like the industry, are certainly benefiting from a strong economy, particularly, in our industries. We continue to see positive audit premiums from our insurers. So the signs that the economy remains strong, I think that bodes well for us. It certainly helps offset the rate declines..

And then the -- Neal, the tax rate outlook for 2020?.

Yes. There shouldn't be any change in tax rate. It's really driven by the fact that how much of underwriting gain we have because that's taxed at 21%. And then how much our investment income is tax-exempt through our municipal portfolio. So it was higher this year, but that was mainly because we had stronger underwriting profitability.

Obviously, expenses affect that as well..

Yes. The state of competition, when we think about the ELCM, Janelle, just looking at the trend here, year-over-year, the ELCM has been coming down kind of 2% to 3% each of the last four quarters. And prior to that, it was 1% to 2% for 1.5 years. So just continuing to drift down low single digits.

If you had to guess, how do you feel like that will shape up given the current nature of the competition?.

Yes. It's still very competitive out in the marketplace. I always try to think when I -- we prepare for these calls, do I think -- do I feel differently about the competition now than I did the last time we all spoke? The answer is no. It's still just competitive out there.

There are still a lot of people that are willing -- actually wanting to write workers' comp and willing to write high hazard workers' comp.

We see -- I guess, we're somewhat -- see some protection in the fact that in our hazard groups, again, A to G, where we see positive results are E, F and G in terms of not only retention, but also binding new business. So that's a positive sign for us.

But let me be very clear, and I've said this in my opening comment, we did not grow policy count or key policy count flat in 2019, which is something that we wanted to do throughout the soft market. We -- I think I've been saying that, at least since '17, maybe even since '16, that's a disappointment to us..

Yes. To that point, again, that you're seeing positive results in the E, F and G. Is that -- say that again..

Yes. And what I mean by positive result is our retention in those classes. And I think I talked about this on the last call, our retention in those classes is high. The new business that we are able to bind is coming in those same classes.

So that's a good sign for us, at least maybe there are a lot more carriers out there wanting to write workers' comp and looking for those premium dollars. But we're -- our high hazard focus gives us some protection from that.

Is that clear?.

Okay. It does.

And would you say that's different? Or is that about the same? Does it feel the same?.

No. I think that's been throughout the soft market. But that's -- I guess, the way I gauge whether is the competition got more intensive, all of a sudden I started losing retention in E that I -- policies I really wanted to retain. Then that would -- that's how I would say, "Oh, there has been a shift in the competition.".

I apologize for going on.

But the -- what has been the -- I don't know if you have it there in front of you, Neal, the cumulative development on '16 and '17?.

I don't have the cumulative development on those accident years in terms of over time. I have a full -- actually, I have a full amount for this year, this calendar year that we've recognized. In '16, for instance, it was -- out of the total, $65 million of favorable development this year. Accident year '17 was $9.5 million. '16 was $23.4 million.

'15 was $14.5 million. '14 was $8.6 million. And then '13 and prior was $9 million..

Then refresh me, was it the first year you took gains out of '17?.

Correct..

We have no other questions at this time. I'd like to turn it back to Janelle Frost for closing remarks..

Thank you for joining us today. February is insurance careers month. Surely, all this talk about trends and challenges aptly highlights the exciting opportunities, a career in insurance offers. At AMERISAFE, our careers are based on turning risk into opportunity. And in 2019, our employees delivered. Thank you..

That does conclude today's conference. Thank you all for your participation. You may now disconnect..