Insperity Inc (NSP) 2006 Q4 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Administaff fourth quarter 2006 earnings conference call. My name is Cindy and I will be your coordinator for today. [OPERATOR INSTRUCTIONS]

Joining us on the call today are Mr. Paul Sarvadi, Chairman and Chief Executive Officer, Mr. Richard Rawson, President, and Mr. Douglas Sharp, Chief Financial Officer. I would now like to turn the call over to Mr. Doug Sharp. Please proceed.

Thank you. We appreciate you joining us this morning. Before we begin, I would like to remind you that any statements made by Mr. Sarvadi, Mr. Rawson, or myself that are not historical facts are considered to be forward-looking statements within the meaning of the federal securities laws. Words such as expects, intends, projects, believes, likely, probably, goal, objective, outlook, guidance, appears, target, and similar expressions are used to identify such forward-looking statements and involve a number of risks and uncertainties that have been described in detail with the Company's filings with the SEC. These risks and uncertainties may cause actual results to differ materially from those stated in such forward-looking statements. Now let me take a minute to outline our plan for this morning's call. First I'll discuss our fourth quarter and 2006 full-year results. Paul will recap the 2006 year and comment on our outlook for 2007. Richard will then discuss recent and projected trends in our direct costs, including benefits, workers' compensation, and payroll taxes, and how these trends impact pricing and gross profit. I will return to provide 2007 financial guidance. We will then end the call with a question and an -- question and answer session.

Now, let me begin by mentioning a few financial highlights. We reported fourth quarter earnings of $13.4 million or $0.47 per share, our highest ever quarterly earnings as we completed a record-setting year. Full-year 2006 earnings per share increased 46% over 2005 on 19% revenue growth. Operating income per worksite employee per month increased 24% to $51 and EBITDA increased $23 million over 2005 to $89 million.

Now for some details on our fourth quarter results. We once again achieved double-digit unit growth, however slightly below our expectations. The average number of paid worksite employees per month increased 11% from 94,031 in the fourth quarter of 2005 to 104,325 this quarter. Paul will provide some of the details on our fourth quarter, 2006 performance and forecasted 2007 unit growth in a few minutes. The 11% unit growth combined with the 4% increase in revenue per worksite employee per month resulted in a 15% increase in revenue over Q4 of 2005 to approximately $353 million.

Shifting to the gross profit, gross profit per worksite employee per month for the quarter averaged $239, above our forecasted range. The big picture for the quarter is higher-than-expected health care costs were more than offset by lower unemployment taxes and favorable pricing. For the fourth quarter, we expected benefit cost to increase 7% over the prior year in order to reach the low end of our annual 2006 forecasted range of 5% to 6%. Q4 benefit costs came in $3.2 million above our expectation, resulting in a 9% year-over-year increase and an annual cost increase of 5.6%. In a few minutes, Richard will discuss some of the details behind this quarter's activity and our expected future benefit cost trends.

Payroll taxes as a percentage of total payroll costs declined from 6.26% in Q4 of 2005 to 5.85% in Q4 of this year. This decline was due in part to a $3.3 million reduction in accrued state unemployment taxes related to our three year old tax dispute with the state of California. In the fourth quarter of 2006, a statute of limitations associated with the rate notice expired and we have therefore determined that an accrual based upon the state's claim is no longer necessary. Workers' compensation costs were 0.86% of non-bonus payroll for the quarter, slightly below our forecasted Q4 cost range of 0.90% to 0.95%. Updated actuarial loss estimates continue to reflect favorable claim trends and resulted in a $2.4 million reduction in workers' compensation costs for changes in estimated losses. This reduction is generally consistent with that reported in recent quarters.

Now, let's move on to operating expenses, which increased 11.5% to $57.8 million for the quarter. We managed our total operating expenses to the low end of our expected range, in spite of further investments in our growth. Salaries and wages increased 22% as a result of the hiring of sales and service personnel in the latter half of the year and additional incentive compensation tied to our improved operating results. Marketing costs increased 15%, as we invested in more advertising and promotions surrounding our fall sales campaign. The investments in personnel and marketing were partially offset by a reduction in both the depreciation and G&A costs of 6% and 7.5% respectively. As for net interest income, we reported approximately $3 million in Q4 of 2006 compared to $1.8 million in Q4 of the prior year. The increase was a result of both higher invested balances and rising interest rates. Our effective income tax rate for Q4 was 33.2% and reflects the impact of additional investments in tax exempt securities throughout the year.

At this time, I'd like to take a few minutes to review our full-year results. We reported 2006 earnings of $46.5 million, an increase of 55% over 2005. Revenues grew 19% to $1.4 billion as a result of a 13% increase in paid worksite employees and a 5% increase in revenue per worksite employee per month. As for the 2006 revenue contribution in growth by region: The southeast region, which represents 9% of total revenue grew by 28%; the northeast region, which represents 18% of total revenue grew by 37%; the central region, which represents 14% of total revenue grew by 26%; the west region, which represents 22% of total revenue grew by 14%; and the southwest region, which represents 36% of total revenue grew by 10%. Gross profit per worksite employee per month increased 5.9% from $221 in 2005 to $234 in 2006. This is reflective of a $3 per worksite employee per month increase in the markup on our HR services from an average of $195 in 2005 to $198 in 2006, and a $10 increase in the surplus generated on our direct-cost programs.

As for our direct costs, benefit costs per covered employee per month increased 5.6% for the year from $590 to $623, while the percentage of worksite employees covered under a health insurance plan increased slightly from 72.4% to 72.7%. Workers' compensation cost as a percentage of non-bonus payroll declined from 1.09% in 2005 to 0.92% in 2006 as we continued to experience declines in both the claim and administrative costs. Payroll taxes as a percentage of total payroll declined from 7.46% in 2005 to 7.27% in 2006. This was due to the higher payroll average of our worksite employees and a reduction in the state unemployment tax accrual related to the California issue mentioned earlier.

Now moving on to operating expenses, which increase 15% on 19% revenue growth. An increase in compensation costs made up about 70% of the increase in total operating expense dollars. And as I mentioned earlier, these costs have been primarily impacted by our decision to accelerate sales and service personnel hiring during the latter harf -- latter half of 2006. Additionally, incentive compensation increased due to our improved operating results and stock-based compensation increased as a result of our annual restricted share grant. In addition to the acceleration of sales rep hiring, we invested in other growth initiatives, including additional marketing investments. This resulted in a 16% increase in advertising costs in 2006. General and administrative costs increased by 8%, below our 13% unit growth, demonstrating leverage in this area.

Now let's review several key balance sheet and cash flow items. Working capital increased by approximately $35 million in 2006 to $128 million at December 31, in spite of the following year-to-date cash outlays: Pay off of the mortgage on our corporate headquarters totaling $32 million, essentially eliminating our long-term debt; repurchase of 614,000 shares at a cost of $24 million; cash dividends totaling $10 million; and capital expenditures of approximately $13 million. These outlays were more than offset by EBITDA of $89 million, proceeds and tax benefit from the exercise of stock options of $29.5 million and reimbursements from our workers' compensation program of $30 million.

Now before providing our financial guidance, I'd like to turn the call over to Paul for his comments.

Thank you, Doug. In my part of the call today, I'll discuss three separate topics. First I'll comment on the outstanding results we achieved in 2006. Secondly I'll provide details surrounding the growth challenges we faced in the fourth quarter and the start of 2007 and our resulting plan of action. Then I will conclude with the discussion of our long-term plans for growth profitability and capital utilization.

2006 was simply an outstanding year for Administaff. We continued double-digit unit growth and achieved the 100,000 worksite employee milestone. We were effective in each of the critical success factors in our business that produce growth and profitability. We exceeded internal targets and set records in sales of new accounts, sales efficiency and pricing of new business, validating the demand for our premium service offering and the effectiveness of our core sales system. We made substantial progress on each of the two most important sales metrics for future growth of the Company; the number of trained sales personnel and the sales efficiency. This year, we accelerated hiring and training of sales staff, which resulted in an 11% increase in the average number of trained sales reps from 219 in 2005 to 244 in 2006. Total sales increased 16% over 2005 on this 11% increase in trained sales reps, as sales efficiency reached a record level for the Company. For 2006, our census to first call rate was over 48% and our closing rate on accounts we bid was over 27%. The sales per sales person per month efficiency rate at our standard client size of 13 employees was 1.29, exceeding our 2005 record level of 1.24, in spite of growing the sales staff.

Our fall campaign sales results were certainly a big part of the year's sales success. For the campaign period September through December, we had a 21% increase in worksite employees sold on a 14% increase in sales staff. Sales efficiency increased to 1.85%, also ahead of last year's record 1.74%. These gains were also accomplished with an increase in pricing of our markup for our HR services on new sales of $7.86 per employee per month or 4.2%, so we continue to see pricing strength on new business. We also had excellent execution throughout the year matching price and cost for the direct cost of payroll taxes, workers' compensation, and benefits, which Richard will cover in a few minutes. Our success in this area produced another record in our most important profitability metric, our gross profit per employee per month, which increased by $13 to $234.

Our operating expense and investment discipline was also a part of our success in 2006. We invested in increasing sales and service personnel, new initiatives, including mid-market and HR tools, and technology development, including a powerful benefits administration system that fits our specific business model. In spite of these additional expenses, our operating expense per worksite employee per month only increased $3, less than 2%. And as a result, $10 of our $13 increase in gross profit per worksite employee per month fell straight to the bottom line and our operating income per worksite employee per month increased 24% from $41 to reach an all-time high of $51.

Now, in spite of these outstanding results for the year and hitting on all cylinders in our core business, we did run into a significant challenge in November and December that effected the starting point for worksite employees in January and created a short-term dampening effect on our growth rate. As you know, we've been working on a new segment in our business we call mid market, which is defined as clients with more than 150 worksite employees, and we focused on this new segment for two primary reasons. First of all, it's a large potential market that no one has figured out in terms of product offering and service model. And secondly, we end up in this business by default when our smaller clients grow into this segment or a core sales rep sells a large account even though our original product and service design did not target this client size.

Our first objective in this segment was to establish a pipeline for sales of new mid-market accounts in order to mitigate the negative impact on our growth caused by mid-market client termination. Over the past year and a half, prior to Q4, we had accomplished this objective. However, since our last conference call on November 1st, new mid-market sales have failed to offset unexpected mid-market client terminations, leading to a 2,500 employee shortfall in our anticipated starting point for paid worksite employees in January. Now due to this shortfall, we are lowering our guidance for unit growth in 2007 from a range of 14% to 16%, which we provided in our preliminary outlook last fall, to a new range of 10% to 11%. Now since this is a significant change, let me provide some details behind how our outlook changed in this short period of time. It's important to note our execution within our core business over this period was substantially in line with our expectations. The route cause of the change in our outlook is the difference between our expectations and our results in the sales and retention in our developing mid-market segment. So let's look at both mid-market sales and retention since our last conference call.

On November 1st, we had a strong pipeline for new business, which we considered possible sales by year end. This pipeline included 26 accounts that were in the decision-making stage, representing about 7,700 worksite employees. Now, although you certainly don't expect to close them all or even a majority of those accounts, our experience would lead us to believe we would close our fair share. Ultimately from these 26 accounts, we sold four accounts. Only one account went to competition, three have declined and kept their HR function in-house, and 18 were unable to decide by year end and are still evaluating our proposal. Now these results are really not that bad, but could have been much better. In addition, one of the four sold accounts -- the largest one with 500 employees -- was enrolled and scheduled for payment in January, but backed out due to the client business being acquired by a larger company.

On the attrition side of the fence, our contract calls for 60-day notice of termination for accounts this size and our experience has been that most of the cli - most of the time clients give us this courtesy. This year, however, this was not the case. On November 1st, we were aware of two mid-market accounts terminating in Q4, one terminating in January, and one in February. And we had a pipeline of mid-market sales from prior periods scheduled to enroll that would offset these losses. However, from November 1st until December 31st, we received notice on seven additional mid-market accounts terminating at or near year end totaling approximately 2,000 employees. Now, the combination of these unexpected mid-market client terminations and fewer new employees enrolled for mid-market sales fully accounts for the 25 employ -- 2,500 employee shortfall in our expected 2007 starting point for paid worksite employees.

Now, there are four questions I'd like to address as a result of this outcome. First, is this systemic issue likely to continue throughout 2007? We believe the answer to this question is no. Although we've had mid-market sales and terminations at other times throughout the calendar year, we believe mid-market sales and retention is even more weighted to year end than our core business. Therefore we believe the risk of recurrence of a number of mid-market accounts terminating mid contract at a time other than year end is low. My second issue is where did these accounts go and do we have a competitive issue? Well, out of the seven unexpected terminations, only one went to a competitor; the other six took the HR function back in-house. When we include all mid-market terminations over last year, the evidence further collaborates that our primary competition is a CFO or senior HR professional leading an effort to bring the function back in-house. We do not believe these results indicate a competitive threat from other providers, but we do need to develop a new strategy to make advocates out of our influencers in these mid-market accounts.

The third issue is where does this leave us on our mid-market initiative and what're we going to do about it? Although we are certainly disappointed with the recent results, we are not the least bit discouraged with our long-term prospect for this segment. This quarter we are in a process of reengineering the mid-market initiative to identify deficiencies and solutions in sales, service, and infrastructure. We will establish our game plan and I'll report on our progress on this throughout the year. The last question I want to answer about this issue is what changes to our 2007 operating plan have we made due to the impact on our model from a lower starting point for unit growth? There are basically two major changes we have built into our guidance we're providing today.

First, we have aligned our staffing plan and incentive target with our 10% to 11 % unit growth expectation. Secondly, we have decided to add two more new sales offices to our plan for the year to [accelerate] the core business strength we have and allow time for mid market to develop. This will bring the total number of new offices for 2007 up to eight. We ended 2006 with 259 trained sales personnel, a 14% increase over the 227 sales reps we had in Q4 of 2005. And our sales person turnover rate is down to around 30% over the last five months compared to previous periods at nearly 40%. Now, if we can continue this acceleration in trained sales staff, it won't be long before we see acceleration in our unit revenue and earnings growth.

This change leads me to our -- my last topic for today, which is our long-term growth, profitability and capital utilization. We remain committed to the compound annual double-digit unit revenue and earnings growth rates similar to those we have achieved over the last ten years in our history as that public company. Since we became a publicly-traded company on the New York Stock Exchange in January of 1997, our compound annual growth rates have been 16% unit growth, 24% revenue growth, and 33% earnings growth. We have accomplished this due to four primary factors.

First, we have an excellent, resilient, and capital-efficient business model. This model includes a recurring revenue stream, built-in pricing strength and substantial operating leverage. Secondly, we have a premium product and service offering delivered with passion that makes a difference in the lives of our small business clients, employees, and families. Third, we have a target market opportunity, including 600,000 of the best small to medium-sized businesses in the United States, and we have a strategy that provides our premium service at an exceptional value to these target clients while generating substantial profit to our Company and our shareholders. And finally, the most important factors driving our past, presence, and future success is our people and our culture. We have a highly motivated team of dedicated people working in an environment that is challenging, rewarding, and recognizes the individual contribution and value.

These factors have carried us through the ups and downs of the last ten years and drove the record-setting results we had in 2006. Last year, we generated $89 million of EBITDA and increased our working capital by $35 million to more than $128 million. In January, our board of directors evaluated our capital strategy and determined we are on track with our priorities. We're in an excellent position to fund our growth, target strategic acquisitions, and return value to shareholders. As a result, we recently increased our dividend and today we're announcing our board action authorizing an additional one million shares to our share repurchase program. We now have approximately 1.5 million shares available to buy back and are prepared to be active in this quarter when our trading window opens.

At this time I'd like to pass the call on over to Richard.

Thank you, Paul. This morning I'm going to comment briefly on the details of our fourth quarter gross profit results, then I will explain how we see our pricing strategy and direct cost trends affect our gross profit per worksite employee per month for 2007. As you know, our pricing model is built by using individual allocations designed to match each of the direct costs, plus a separate allocation for our HR services, we call the markup, that is designed to generate most of our gross profit per worksite employee per month. Since all of the direct costs are not known in advance, we build in targeted allocations to cover those particular costs. Our pricing strategy is to have these allocations match or exceed the direct cost and produce a surplus of approximately 3% that when added to our markup produces the desired gross profit per worksite employee per month.

For this quarter, as Doug just reported, our gross profit per worksite employee per month of $239 was above the top end of our forecasted range by $1 per worksite employee per month. These results came from achieving our targeted average markup for the quarter of $200 per worksite employee per month and generating a surplus of $39 per worksite employee per month, or 4.1% of total direct costs allocations. This surplus came from contributions in two of the direct cost centers offset by higher-than-expected cost in one cost center. Let me explain.

The surplus in the payroll tax cost center has continued to come from better-than-expected unemployment tax rates that we received earlier in 2006 compared to our pricing allocations. Additionally, the reduction in expense from the state of California rate issue, that Doug mentioned a few moments ago, also contributed to this surplus. Our second contributor to this surplus in this quarter came from our workers' compensation program. The better-than-expected surplus came as a result of managing the settlement of workers' compensation claims at lower-than-expected costs. In addition, we continue to see favorable trends in the incidence rate on the current policy. That rate is down 2% on a worksite employee base that has grown 11% this quarter.

That leaves the benefits cost center, which we -- which did not contribute to gross profit surplus at all. Specifically, the healthcare cost in the quarter were $16 per covered employee higher than what we had forecasted for the quarter. This spike was driven by a 35% increase over prior quarters in both the number and dollar amount of large claims that exceeded $50,000 per claim. Over the years, United Healthcare has told us that our experience with large claims has generally been lo -- at the lower end of the range of their actuarial expectations. This quarter we were at the high end of their expected range. Now, despite this unexpected increase in the fourth quarter, the plan experienced a 5.6% increase for the full year, which is substantially lower than the cost trends seen in the small business marketplace.

In summary, 2006 was an exceptional year from a pricing and direct cost management perspective. Our average markup increased $3 per worksite employee per month to our target of $198 per worksite employee per month for the year. And our surplus increased from an expected $24 per worksite employee per month for the year to $36 per worksite employee per month. It was a job well done by a lot of great people.

Now, I'd like to update everyone on the changes to our gross profit per worksite employee per month expectation for 2007 since our last conference call. The bottom line is, we have a wider range in our expectations for this metric than we did last fall to account for a wider range of possible benefit costs. Let me explain each component beginning with pricing. Our plan this year is to increase the average markup from the 2006 average of $198 per worksite employee per month to $201 per worksite employee per month for 2007. The demand for our services is still very high and renewal pricing in the fourth quarter increased slightly more than $4 per worksite employee per month.

The second component of our gross profit per worksite employee per month comes from the surpluses that we generate by managing the direct cost so that they are lower than what we allocate in our pricing to cover those costs. Historically, we generate a surplus from our payroll tax cost center and 2007 should not be any different. Our preliminary guidance last fall included estimated state unemployment tax rates from Price Waterhouse Coopers consistent with how we budget this expense every year.. Now, most of the states have given us our unemployment tax rates for 2007 and the expected costs will be lower than 2006 and somewhat better than we anticipated last fall. Because of how payroll tax expense and the corresponding billing allocation work in our model, we should see a nice surplus in Q1 followed by nominal surpluses in Q2 and Q3 and then a moderate surplus in Q4. As for workers' compensation, we see a surplus contribution continuing throughout 2007 generated by the difference between our allocations and the ultimate cost of the program. We expect our cost to remain at the current level in a range of 0.85% to 0.90% of non-bonus payroll for each quarter of 2007. This range is also slightly better than we had expected last fall.

Now the benefits cost center is the driver to the change in our expectations for 2007. Last quarter, I spoke at length about our cost expectations for 2007 and the reasons why we would expect our benefit cost per covered employee to increase 7% to 8% over 2006. The new factor to consider today is the higher-than-expected large claims we experienced in the fourth quarter. We believe the conservative approach is to include both possibilities related to these large claims. If that trend were to continue throughout 2007, we could see our cost per covered employee increase 9% over 2006. If the increase in large claims is not a trend, then our 7% low-end expectation is still valid. Therefore, our new range for expected costs is a 7% to 9% increase per covered worksite employee over 2006. Since our current pricing allocation is based on a 7% to 8% increase, we could have a shortfall that would translate into a $3 to $4 per worksite employee per month reduction at the gross profit line. We should know in less than 90 days whether Q4 was a blip or a beginning of a trend. If it is a trend, then we'll have to add further increases to our allocations. If it was a blip, then we will not be increasing those allocations as aggressively.

In summary, as we look to 2007, we see the average markup per worksite employee per month to gradually increase throughout the year from an average of $198 to $201. The surplus generated from two of our three direct cost centers is better than what we thought last quarter and the benefits cost might be worse if large loss claims continue. Therefore, the surplus component of our gross profit for the full year could be $26 to $35 per worksite employee per month.

Now, I would like to turn the call back over to Doug.

Thanks, Richard. As most of you are probably aware, we provided our preliminary 2007 outlook during our previous conference call. This outlook was provided prior to the completion of our fall sales campaign, year-end client renewal period, and finalization of our operating expense budget. So at this point, I'd like to update our full-year guidance and provide some detail by quarter. Let's begin with worksite employees. As Paul mentioned a moment ago, when you take our January starting point, combined with our forecast for net unit growth of 1,100 to 1,300 for each month thereafter, we are now forecasting 10% to 11% unit growth for the full year of 2007. As for the first quarter, we now expect the average paid worksite employees to be in a range of 105,500 to 106,000.

Now let's discuss our pricing and direct cost metric gross profit per worksite employee per month. As Richard mentioned, our 2007 forecast has been updated since our previous conference call. We now expect full-year 2007 gross profit for worksite employee per month to be in a range $227 to $236. The low to the high end of this range is primarily based upon the expected range of outcomes of our healthcare claim costs. We plan to adjust the pricing of the healthcare component of our fee based upon this range of expected cost. The price increases will take effect as clients renew and will large -- and will therefore largely impact gross profit favorably during the latter half of the year. So as for the quarters, we are now forecasting a range of gross profit per worksite employee per month of $224 to $230 in Q1. We would expect Q2 to decline by about 1% to 2% from Q1 due to increased payroll tax cost associated with new business. Thereafter, we would expect a sequential increase of 4% to 5% in Q3 and again in Q4, as payroll tax costs decline and price increases take effect.

Now, let's discuss operating expenses. Since providing our preliminary guidance of a 13% to 14% increase in 2007 annual operating expenses, we have finalized and recalibrated our operating expense budget in light of our revised unit growth and gross profit guidance. We are now forecasting an 8.5% to 10% increase in operating expenses, or a range of $240 million to $243 million. The high end of our full-year range is tied to achieving higher unit growth and gross profit goals. Under either scenario, we are targer -- targeting a decline in operating expenses on a per worksite employee per month basis from $183 in 2006 to approximately $181 for 2007.

As for the details, we expect an 8% to 10% increase in cash compensation costs over 2006, with the upper end of this range tied to additional incentive compensation upon achieving higher unit growth and gross profit results. Cash compensation costs will increase sequentially each quarter throughout the year with the hiring of sales reps as we open new sales offices. Stock-based compensation costs are estimated at $7.1 million and step up in the second quarter with the annual restricted share grant scheduled for March and the board of directors grant in May. We expect depreciation and amortization expense to remain relatively flat when compared to 2006, between $14.5 million and $15 million and consistent from quarter to quarter. As for commission expense, we expect an increase between 12% and 13% over 2006, slightly higher than our forecasted unit growth due to the change in the compensation structure mentioned in our previous conference call. These costs should increase sequentially over the year as our worksite employee base grows.

Advertising expense is expected to increase approximately 2% over 2006, and because this is a nonelection year, we can increase our air time without a corresponding increase in expense. As in previous years, these costs step up in Q2 and Q4 due to advertising associated with our spring and fall sales campaigns. General and administrative expenses are expected to increase approximately 8% over 2006 and are higher in Q1 due to costs associated with our annual sales conference and employee incentive trip. So as for the first quarter, we expect total operating expenses to be in a range of $58.25 million to $59 million. We expect net interest income to be between $3 million and $3.2 million for the first quarter and between $13 million and $14 million for the full year, based upon forecasted cash balances and current interest rates. We are forecasting an effective income tax rate of 36.2% and expect average outstanding shares of 28.2 million for the first quarter and for the full year. It is important to note that our forecast of net interest income and average outstanding shares excludes the impact of any share repurchase activity. And finally, we have budgeted capital expenditures at $15 million.

So at this time, I'd like to open up the call for questions.

Thank you. [OPERATOR INSTRUCTIONS] And your first question will come from the line of Tobey Sommer of SunTrust Robinson Humphrey. Please proceed.

I was wondering if we could dig into the middle market clients who left in November and December? Was there any additional color you could give us in terms of any industry concentration or geographic concentration that would give us a little bit more insight into what happened?

Well, Tobey, there's really nothing further beyond the -- nothing geographically. There were only seven accounts, but they totalled the 2,000 worksite employees and it was just basically companies making a decision to move -- as I said just mostly taking it in-house at year end.

In taking a look at the 10% to 11% growth in worksite employees in '07. if you look at the core traditional customer base, what sort of growth rate is implied in that 10% to 11% for the core smaller customers?

Yes, I think your -- I don't have that specifically, but it's higher than that because we're basically want to be conservative about giving ourselves time for the mid market to develop. And I think it's prudent at this point to factor in that and be certain that your -- you factor in what we just saw, so that's what we did. We want to be conservative about the going-forward growth and then put people on to the -- both the mid-market game plan and accelerate the core market selling program and see things accelerate as we move on later in the year.

Okay. Another question for you regarding the -- not necessarily the service fee -- the markup or the surplus, but over time we've talked about you finding other areas that can contribute both profit to that gross profit per worksite employee per month. Do you have any comments on what sort of expectations we should have for that expanding over time?

We really didn't break that out very much in this particular call because we had so much to cover. But last quarter we talked about the fact that we do have a couple dollars of gross profit per worksite employee expectation in this year's model related to the total of the other initiatives, which includes HR tools, which is selling other HR services and products to prospects as an introduction to Administaff and the quality of our services, try to bring in more prospects that way to our core business. And also we have the marketplace, which is -- consists of other 50 high-quality strategic alliances where we sell other goods and services to our current client base and their employees and families. So both -- between those two major initiatives we expect to see that begin contributing this year and we think that has tremendous potential for growth over the years. And we look forward to demonstrating that over the course of this year.

Last question and I'll get back in the queue. Do you have any expectation for the number of trained sales people that you're targeting for being in place for the fall campaign in '07?

Yes, we do. This past year, we did a great job of getting back on the right end of growing the sales staff with the 14% increase by year end. And our broad based initiatives that we looked at to effect turnover rate have taken hold. Over the last five months or so we've seen the turnover rate go down to 30%, which is really exciting. You were starting to see more people move into that tenured position of more than 18 months with us and that's where the efficiency rate really grows. Over the course of this year, we're opening eight new offices and so you can kind of build in about an average of maybe five, we're -- which is 40. Takes us from about the 259 or 260 to 300 as an average. We get to that number 300 or so by fall campaign. That's where we'd like to be.

And that's -- 259 was the hired number at the end of the fourth quarter o r was that the train number?

Trained number.

Thank you very much. Thank you.

And your next question will come from the line of Mark Marcon from Robert W. Baird. Please proceed.

Morning. I was wondering, how big is the mid-market segment for you right now just in terms of percentage of worksite employees?

Yes, that's a good question. The mid market for us at this stage of the game, after we've seen some deterioration over the past quarter in that market, is down to 39 accounts and a total of right around 11,000 worksite employees, kind of the good news and bad news. The bad news is it's down to that, the good news is it shouldn't be able to affect us too much more going forward before we see -- start to grow that segment again.

Is the -- is the fee in the GP per worksite employee lower for those clients than it is for your average client?

Yes, it is. If you'll recall our gross profit per employees runs about 10%, a little over that. Maybe 10% or 11% lower than the rest of the book of business. But the operating expen -- operating income per worksite employee we think is probably just a little better than our core business.

And is that -- is that -- 11,000 worksite employees, is that, how much is that down year over year?

Well, the peak I think was around 14,000 worksite employees at one point, so that kind of gives you a framework of where that is over that period of time.

Okay. And why are they -- why are these mid-market employees bringing it in-house? When you talk to them, what's the primary driver?

I think it varies kind of from account to account. But I'll tell you, a lot of them -- what I pointed out in our script that you have a new person involved, either a CFO or a senior HR person, and a lot of times they're either staking out their claim or trying to make a cost-related impact and they weren't involved in the original agreement with our Company or in the execution of our game plan. But I think the bottom line is we haven't in our process developed the right relationship at the right levels in the client company to mitigate that when a new influencer comes into the mix, so that's something we'll be working on.

Okay. And how's retention with your smaller core clients?

Very good. It was excellent throughout the year, throughout the fall campaign period. The whole situation we looked at from Q4 to where we are today is this mid-market level.

Okay. So the retention for the smaller core clients, that remains in the 81% range?

Yes. In fact, the actual -- we kind of talk about an 80% range on client retention. We actually had that. For last year, our total client retention would be right around 80%, but the ones that went away were larger -- on average, larger accounts because of these mid-market accounts and that's why you have that effect at the worksite employee level.

So when you talk about reengineering the mid-market initiative, is it primarily in terms of the client contact or is there something that's going to change with regards to the service that's being provided or the pricing or --?

We're -- when I say reengineering I intend to convey that we're not going to tweak this. What we're doing at this point when we'll complete this quarter, we're taking each group. The sales -- mid-market sales group, we've already had initial retreat with that group to do a swat analysis -- strength, weakness, opportunities, and threats -- and have a brainstorming session. We're going to do the same thing with our mid-market service team and the same thing with what we call our infrastructure -- mid-market infrastructure group, which effects the areas of benefits and legal and finance, and that group that provide the infrastructure to these two other teams that sell and service the accounts. We're going to have that retreat with each one of those groups separately, then bring the group back together and nothing is off the table in terms of pricing. in terms of how we sell it, how we service. At this stage of the game, we've learned a lot, but it's time for us to bring all that together and then have a real step up of effectiveness in this area.

So it sounds like once you've completed that step, your guidance could change again, could it not?

Oh, it certainly could.

Okay, great. I'll jump back in the queue.

Thank you.

And your next question will come from the line of Jim McDonald of First Analysis. Please proceed.

Yes, good morning, guys.

Hey, Jim.

Morning, Jim.

I just wanted to follow-up on a couple things here. In terms of your metrics, do they exclude the mid market or how do you handle that in your metrics? And when you said 80% retention, was that a number of customers or of worksite employees?

Yes, see the 80% client retention is number of customers. We had more employees go away at that level than we did before, of course, because first of all the customer base is bigger, but also the fact that the average sized client that went away was higher than last year. That's -- that's the answer to that question. Now what was the first question again?

The first question was when you give your metrics that you're selling X percent of budget, do you somehow exclude mid market there or are those included in that sales?

No, it's all included.

Okay. And then going back to the metrics question, so what would be retention be for a worksite -- on a worksite employee level?

I didn't really work out the numbers that way. But again, this is what we're digging into kind of by segment. Obviously we just discussed the fact that you're actually down in the mid-market segment and grew faster than our average rate in the rest of the business. This is -- it's really what we feel like is a very isolated incident related to this particular year end and that 2,500 employee difference, but we will be digging in on that going forward.

Well, I think the other thing to remember is that in the whole rest of our book of business, the customer satisfaction survey results were record setting for 2006 in terms of -- in terms of the survey results that we had, so it's not like people are abandoning the ship here. They love the services and they're willing to pay more for it every year.

Yes, we really drill down on every aspect of the core service side of the business and core selling side of the business, and we're continuing to hit on all cylinders in that regard.

Okay. You mentioned that everything's on the table. What about the possibility of offering mid-market clients on a non-PEO basis?

I think in order to do an effective brainstorming, exercise, you can't have limitation. I'm going to say if the amount we invest is no object, if -- we're going to take everything off the table to brainstorm. That doesn't mean that's where we're going to land, but I don't think you can have an effective brainstorm if you don't put that option on the table. But that's -- that's not -- at least in the first round -- the first round of retreat I had with the sales group, we know that the coemployment models bring in some substantial advantage to our mid-market customer. And so I would look for that as a likely outcome, but we're certainly going to explore that as part of potential solutions.

And let me just change gears to the benefits for a second. I may have missed this number, but did you give your benefits cost for healthcare in the fourth quarter?

Yes, it ended up being $646 per covered employee.

Okay.

Not worksite employee, but covered employee.

Right. And just in terms of dealing with some of the larger losses you've had in that area, are you doing anything different on underwriting? Are you looking to -- is this concentrated in several accounts that obviously you could drop? Or is there a possibility of reinsuring any of these large losses?

Well, you've asked a number of questions of which we've all asked, as well, so let me just try to answer it generally for you, Jim. Specifically, this quarter there were just a random nature of how healthcare system works, and there's some things that you can control and some things that you can't completely control. And large loss claims you bake in an estimate every quarter for what you think they're going to be. And as I in my prepared remarks for 2006, we've been rocking along within a few dollars of a number of large loss claims every quarter have been virtually the same, even though we've been growing, so this fourth quarter pop in those -- or spike in those was certainly not expected. I can tell you that we do get the breakdown of claims by locale, by local, by the plan of which they came in. We get that kind of information by customer. We get it by the type of -- the type of claim, like was this a premature baby or did someone have a heart attack or whatever. And so we're constantly analyzing that information to see if there's things that are sys -- that become systemic to the driving our health plan costs. And if we find that there's something systemic, then we can make a change. If it's kind of random, then there's not much you can do.

And your next question will come from the line of Cynthia Houlton of RBC Capital. Please proceed.

Hi, just, first on a clarification. You said the 30% attrition that you're seeing in your sales force, that's a -- over five months that is an analyzed number, right? So in the last quarter you mentioned it was in the high 30s. This quarter it's gone down that much? Is that correct?

Right. It's an annual turnover rate that's down to around 30% from what was running at high 30s, near 40%. And in historical perspective, the highest we've ever been is kind of in the low 40s, and the lowest we've ever been is where we are right now.

And in terms of compen -- annual bonuses, is that something that's paid out in the March quarter? I want to get a sense of timing if that's somewhat due to timing of bonus payments.

Good question. From a compensation standpoint, we really don't have a big bonus program for the sales people that has people hanging on until a certain time. In fact it's quite the reverse. By this time of the new year, we usually have a culling related to sales staff that maybe were not successful in the fall campaign. In other words they actually get their increase in compensation by selling more through the fall campaign. And then if they didn't make it through that period, that's when our managers would say, look if they can't make it during our peak selling period, they're not going to make it in the rest of the selling period. And things look really good right now. We don't have what we've seen in historical periods of a significant number of sales staff that we're terminating for lack of production from the fall campaign. That further validates that in the core business we had a very strong fall campaign and it was very much across the board.

And then just a follow-up on mid-market initiative. Could you put some numbers around how many people are dedicated to that effort now? Meaning that, again, is this a fairly large group currently? Is this a risk that this group maybe gets reduced significantly or just trying to get my arms around how many people are dedicated to mid market right now?

I think that's a real good question. Let me tell you part of the answer, I think, is part of -- is part of where we could have done better in getting to this point. What I mean by that is is we took somewhat of what I would call incremental approach to mid market. We have a total of about ten or 12 people in mid market sales; eight or so salespeople and then a support staff. On the service side of the fence, we have quite a few people who touch servicing mid-market clients, but it's actually pretty hard to even break them out because they're a part of our entire service organization. There are service teams that are out there in the regional processing centers and then also individuals that are out in the service offices, the local offices.

So you have people spending part of their time on mid-market accounts and part of the time on our core service account. That incremental approach, I think, is part of maybe why we haven't progressed as quickly, because you don't customize as much to fit specific needs if you don't have it segmented off as effectively. So we'll look at how we're organized in the area and some of those more basic things to get structured to make more improvements, but don't have a direct answer to the question. But even if those people that are involved in that effort, I wouldn't see us reducing that group. We would be redirecting them to others -- other -- to our core segment or really seeing this thing ramp up and maybe segment them off into a different service unit.

And your next question will come from the line of Michael Baker of Raymond James. Please proceed.

Yes. I was wondering if you could comment in general on the slow down in growth in the west region. Is that primarily related to a concentration of these mid market defections in that area?

No, actually it wasn't. It's really more related to the number of trained rep count by region as that compares to the base of our customer base in each region. And that continues to be the main driver of our regional results. In other words, where we put the sales staff is where you get the sales, and you've got to keep the number of sales people up high enough for the size of the base. And what happened is we grew the west coast very quickly, especially down in southern California. But we had some -- we had some office -- offices that weren't producing like they should. And when you have 42 offices, you always have one or two or three district managers that maybe aren't getting the job done. And we had some of that happen both in Phoenix and San Francisco. So we replaced those managers and even replaced the regional manager, so you actually had a reduction in the number of trained reps, did a little house cleaning out there. And over the course of last year, you went down from in the mid 50s to mid 40s the number of trained reps on the west coast. And so that kind of ultimately flows through the model and you see a little slower growth out there.

Were there any competitive factors that have changed, particularly in the California market that's influenced anything or is it pretty much just what you talked about?

It's just what I just said.

Okay, and then second question. If you do have to increase health benefit pricing, can you just give us a general sense of how sensitive your clients are to that?

Yes, I think -- I was just going to say that -- I'll let Richard comment on that that's in his area -- but we don't see any pricing scenario outlook for us that puts our pricing on this component even at the kind of rates that small businesses are seeing increases. We feel pretty good about maintaining our competitive advantage there.

That's helpful. And then finally, a number of states are starting to put forth some proposals to try and deal with the uninsured. Are there any specific provisions that you either view as a benefit or a detriment along those lines?

Actually I was just going to say any time that there's a regulatory environment that's starting to mandate something like in the area of our direct cost, we've always seen that as a positive, not a negative.

You have to be sure that you're included in the specific initiative and that you don't get hurt, which we did, for example, up in Massachusetts when they put their proposal together. We had our people on the ground there making sure we were in good shape there and we're fine with that proposal. And if that extended across the country, I think we'd be in real good shape to be one of the kind of aggregators that they look at in that model.

The other development on this issue, though, that is really exciting to me, I want to keep myself from being too excited about it because you don't -- until it happens, it hasn't happened. But as you all may be aware, the Senate passed a wage and -- I'm sorry, a minimum wage bill a couple weeks ago. And in it was a package of small business tax breaks, if you will, that would help small businesses offset a minimum wage hike. We've been in Washington for maybe 15 years trying to pass some type of certification-type legislation at the federal level that would would recognize our industry and offer the opportunity for us to become certified and obtain what's called successor status for the purpose of certain taxes that would ultimately lower the cost of our service for our customer and increase our profitability. So it's kind of a win-win, because right now the government gets a double dip when a customer comes into a PEO relationship, the wage basis start over and we pay more tax for really no good reason. But any way, we actually were in that minimum wage bill in the Senate. Our provisions were part of that small business package.

Now, the House has since passed their version, which also turned out to be a minimum wage hike and a much smaller package of small business tax incentives. We were not in the House version, so that has to go to conference committee and who knows what happens there. But if we survive that, that would be a huge win for our industry and it's something our Company's been working toward for, like I said, a long, long time.

And your next question will come from the line of Thomas Giovine from Giovine Capital Group. Please proceed.

Hi, guys.

Hi, Tom.

Hi, Tom.

Quick question. First is one of the things I think you sort of do yourself a little bit of a disservice is the -- the attrition due to middle-market clients that were as a result of M&A, what exac -- how much was that? I guess my point was that you should probably break that out, because it's really completely control of -- IBM buys one of your bigger clients.

Yes, and we have. That's that old -- we have a success penalty, if you will, in a couple of different ways. Companies grow and they're successful and then they get purchased.

But what was the actual number this quarter?

We had -- let me see. I don't have the exact number. I'll get that for you, but I know of two for sure that were sales --

So that was about how much?

-- without researching it. But I'd have to go back and get you a hard number.

That was what, at least 200 employees, 300?

Oh, no. The two -- one was 500.

Right. Maybe an amount of what folks are sort of concerned about I'm assuming at this point?

Yes, and I think that's fair. I can go back and get that information. I think that -- remember, the first goal of our whole initiative was to have big enough pipeline to kind of offset those because they're going to happen.

Okay. And then I have a couple of quick mini questions here. First thing is, sort of refresh my memory -- and I apologize for -- this is sort of elementary for a lot of people, but the -- if I look at cash, restricted cash and marketable securities, it comes out to be about $9 a share. What of that, again, is sort of temporary and what exactly is restricted cash? In other words of that $9.25 per share, what is that -- as a shareholder can I look at as pure real cash, [multiple speakers] temporary factors or whatever?

Yes, the restricted cash is dealing with moneys tied up in our workers' compensation program. I think it's best we steer our investors to look at working capital as far as our liquidity and that we ended the year at about $128 million in working capital, and that's a good number for you to look at.

Okay, so $128 million. So I could think of it like at 128 --

$6 bucks a share.

About $6 a share.

Right.

Great. And I guess my other question is that I don't know whether -- and this is maybe -- I won't try to pontificate for too long -- but if you look to see what's happening in the market, there's clearly a lot of arbitrage going on between the public and private market's tolerance for debt levels. I was sort of disappointed to see you pay off your debt for the building because debt's sort of a good thing, and if you think about you have public companies with three times debt to EBITDA, which would mean you could put on $265 million worth of debt easily. In the private market, they're putting on ten times EBITDA.

And I'm not advocating you do that, but the thing is at 5%, I'm not that interested in Administaff having a cash management business as much as I am seeing you get a little bit more aggressive with a buy back or a dividend, because the fact that, as you mentioned, you don't use -- you're not a big cash user as you grow. In fact, you generate positive working capital. I'm just trying to understand why we're not being more aggressive with either a dividend or a share buy back. If you subtract $6 of cash, $29 and then whatever, maybe your earnings will drop a little bit. But you're still spending -- on what earnings estimate is, you're still turning around 14 times earnings, which is ca -- by buy back stock it's massively accretive on a cost of equity basis. Why aren't we seeing a little bit more aggressive buy back? There is a little bit of share creep and --.

Yes, I think that's --

Again, I rambled a little bit, but I'm just trying to understand what the thinking there?

Yes, that's a really good question. In fact, that was the primary discussion of a two-day board meeting we had in New York just a couple weeks ago. We brought in some banks to kind of give us their view of our capital structure and use of capital going forward and just to get some input from others. And where we landed is very much what you're describing. We increased the dividend 22%, I believe, immediately. We then, as we announced today, increased the authorization by one million shares, which gives us 1.5 million shares authorized to buy back. Obviously, we have plenty of cash to be aggressive buying shares back and anywhere in the range, even where we started today, is certainly accretive. So I expect us to be active buying our shares back. We've fully investigated the acquisition pipeline that we're interested in and what requirements the business has to accelerate our growth. And there's plenty of funds left over to buy shares back and that is very much accretive to us and I expect that to be part of our game plan going forward.

And your last question will come from the line of Jim Wilson of JMP Securities. Please proceed. Mr. Wilson, your line is open.

Well, I don't hear Jim on the line.

Sir, that will conclude the question and answer portion of our conference call. I will turn it over to you for any closing remarks.

Once again we want to thank everyone for being on the call today. We look forward to seeing you out at the many conferences that we'll be involved in this spring and then back with you next time around. Thank you, again.

Ladies and gentlemen, thank you for your participation in today's conference. This does conclude the presentation and you may now disconnect your lines. Everybody have a wonderful day.