Good day, ladies and gentlemen, and welcome to the Fastenal Company 2017 third quarter earnings conference call. At this time, all participants are in a listen-only mode. Later, we will conduct a question and answer session, and instructions will be given at that time. If anyone requires operator assistance, please press star then zero on your touchtone telephone. As a reminder, this conference call may be recorded. I would now like to turn the conference over to Ellen Trester, investor relations. You may begin.
Welcome to the Fastenal Company 2017 third quarter earnings conference call. This call will be hosted by Dan Florness, our President and Chief Executive Officer, and Holden Lewis, our Chief Financial Officer. The call will last for up to 45 minutes and will start with a general overview of our quarterly results and operations, with the remainder of the time being open for questions and answers. Today's conference call is a proprietary Fastenal presentation and is being recorded by Fastenal. No recording, reproduction, transmission, or distribution of today's call is permitted without Fastenal's consent. This call is being audio simulcast on the Internet via the Fastenal Investor Relations homepage, investor.fastenal.com. A replay of the webcast will be available on the website until December 1st, 2017, at midnight Central Time. As a reminder, today's conference call may include statements regarding the company's future plans and prospects.
These statements are based on our current expectations. We undertake no duty to update them. It is important to note that the company's actual results may differ materially from those anticipated. Factors that could cause actual results to differ from anticipated results are contained in the company's latest earnings release and periodic filings with the Securities and Exchange Commission. We encourage you to review those factors carefully. I would now like to turn the call over to Mr. Dan Florness.
Thank you, Ellen. Good morning, everybody. Thank you for joining us on our third quarter earnings call. Holden's asked me to stay pretty tight to the script today. I'm going to be going through the slide deck, first couple slides of that. I will preface my comments with a few thoughts. This is as much a shout-out to our Fastenal team as it is to our Fastenal shareholders. My first comment to the team is, job well done in the quarter. For nine consecutive months in 2017, we have exceeded our internal goals for sales growth. I'm proud of the team. I'm happy for the team. Last December, I challenged them at our annual gathering of the leadership of Fastenal, with three simple messages as we approached 2017. One is, let's go out and grow our sales. Let's get the business growing.
We have a lot of growth drivers. Let's execute and grow our sales. Secondly, let's manage the business well and grow our earnings, because long term, the rewards we can create for our customer, our employees, our shareholders, and ourselves are centered on our ability to grow our earnings and our returns long term. If we grow it faster than other companies that you can choose to invest in, the market will reward us from the standpoint of valuation, and our employees will benefit from that because a fair number of our employees are also shareholders of the company. Finally, I challenged them to think big about the business. We're going through a massive transformation in the business right now. We've talked about it in previous investor days. We've talked about it in our filings.
In the last 10 years, we've grown a tremendously supportive business within Fastenal called Vending. In the last three years, we've grown a tremendously supportive business to Fastenal and to our customers called Onsite. They're complementary to each other. They're also businesses that Fastenal is uniquely qualified to be successful in and to think big about our future and how the business is going to morph. As you noticed in the release, our sales growth accelerated in the third quarter of 2017. Year to date, our daily sales growth is 10.1%. That's the first double digits we've seen since 2014. As I called out in the quote in the release, we ended the quarter with 15.3% daily sales growth in the month of September. Again, very pleased with top-line performance in the quarter. Our growth drivers continue to gain traction.
We had record signings in the third quarter as it relates to Onsite, putting us on track for a very aggressive goal of 275-300 signings. Recall, if you go back four years ago, five years ago, six years ago, on average, we were signing about 10 a year. In 2015, we began to move the needle, and we began to challenge our team to sign more. We improved the number to about 75, about 175 the next year. Our goal this year was to add another 100 again and do at least 275. I feel very good about us being on pace to accomplish that. Year to date, we've signed 213, 64, 68, and now 81. In all that, we've been investing behind the scenes in additional infrastructure to support the Onsite business. That includes IT infrastructure. That includes implementation infrastructure.
Despite those investments, we've continued to obtain operating expense leverage in our business. Our incremental margin was 21% in the quarter. One thing about the quarter, it's a 63-day quarter versus a 64 last year. For us to lose a day in a quarter is kind of unusual. We do just over $18 million a day in business. If we would've had a more normal quarter, 64-64, most of that incremental gross profit dollars would have gone to the bottom line, and our incremental margin on that business would have been about 25%. Feel very good about the expense management we're exercising as we grow and invest in the future of our business.
If I look at it from the standpoint of how much of that converted into cash, our operating cash flow in the third quarter was a record for any third quarter in the history of Fastenal, both from the standpoint of absolute $ and as relative to our earnings. Very pleased with the team, and that requires a big effort from both our supply chain folks as well as our branch and Onsite personnel because of managing the inventory growth, because that is the biggest wild card when it comes to our working capital needs. Touching on the Onsites again, going to the second slide. 81 Onsites signed. We have about 555 active sites. That's up almost 48% from a year ago. Again, I reiterate, our goal is 275 to 300 signings. A sidebar I'll throw in as it relates to Onsite.
As we've ramped up our Onsite in the last several years, it's translated into our ability to grow and take market share faster. In yesterday's meeting I had with our board of directors, I was walking through with them that concept of the growing sales component. Last year, in the third quarter, about 25% of our district leaders, so we have about 250 district leaders in our business, about 25% of them grew their business double digit. It's no coincidence, or it is a coincidence, that in the prior year, in 2015, about 25% of our district managers signed an Onsite, and that translated into 75 signings. Last year, I've shared this on calls before, last year, we more than doubled our signings because roughly 54% of our district managers signed an Onsite in 2016.
Interestingly enough, in the third quarter of 2017, 54% of our district managers grew their business double digit. Year-to-date in 2017, our goal for the year in establishing the 275 to 300 Onsite signings, our goal was to keep improving the participation of our leaders in this growth driver. I challenged the team, can we get to 80%? We'd gone from 25% to just over 50%. Can we add another 30% and get to 80%? Year-to-date, 64% of our district managers have signed an Onsite. Success for us in Vending 5 and 6 and 7 years ago came from more participation across the organization. Success for us in Onsite follows the same path. A greater portion of our business engaged with their customer, and the manifestation for us is Vending and Onsite signings. Year-to-date, it's 64%.
I don't know if we'll accomplish 80% by year-end, but I'm really impressed with what we've done in the first nine months of the year. Total end market sites today, 2,973. A year ago, we were at 2,921. That continues to grow as we morph a piece of our business closer to the customer. As you see from our filings, we've closed some branches in the last 12 months, as we've done in the last 5 years. Think of it as not a closing, but think of it as a consolidation. We've consolidated 2 branches because part of the business of one has moved Onsite, but our local presence continues to grow. We signed 4,771 Vending devices, basically on par of what we did in the third quarter of last year. To me, the most meaningful thing is our business there continues to grow double digits.
It was just over 20% growth, 21% on a daily basis, just under 20% on an absolute basis, because one less day. Our number of devices we're removing inch down as we've gone through the year and inch down in the third quarter. Very pleased from the standpoint, it feels like our signings are better and our performance on the existing base is better. Finally, our national accounts business grew about 17% in the quarter, which means our non-national business grew just over double digits for the first time in quite some quarters to give us our combined growth. Very pleased with the quarter. Our growth drivers are moving the needle. We're managing our operating expenses through that. One thing you'll notice, I've been completely silent about our gross margin. Holden's going to touch on a few points of that.
I'm sure there will be a question or two on gross margin, but I'll throw one piece out to the group in the for what it's worth category. If you look at Onsite, if you look at Vending, and if you look at our Mansco acquisition, add these three components of our business up, and they account for about two-thirds of our growth from the third quarter of last year to the third quarter of this year. We've previously said all three of these operate at a gross margin below our company average but have very attractive operating margin and return characteristics. At the end of the day, that's what you pay us for. With that, I'll turn it over to Holden.
Thank you, Dan, and good morning. Before jumping back into the slide deck, I did want to call out a few items that made the third quarter of 2017 unique. The first is, I just wanted to call your attention to the last page of the supplemental deck. This quarter, we did take a fresh look at the market exposure of our business and made certain adjustments that affected our end market mix and slightly altered the growth rates versus what we've reported in the past months and quarters. Please feel free to take a look at that last page. Secondly, as Dan indicated, we did have one less selling day in the third quarter than was true the prior year. Based on our September daily sales rate, that cost us more than $18 million in sales during the period.
That was known, of course, keep in mind that many of our costs, notably the employee-related expenses, are not days dependent, and that does affect leverage. Third, we estimate that the hurricanes that hit the Gulf, the Southeast, and Puerto Rico shaved 20-30 basis points from revenue during the quarter. It's hard to estimate the impact of post-storm recovery on our volumes, but we do believe that some of that was present in September, particularly in the Gulf, though not at a magnitude to offset the original lost sales. In addition to the top-line impacts, the hurricanes also were a very slight drag to gross margin due to the write-off of damaged assets and an increase in sales of low-margin storm-related products like water and generators. With that said, let's look back at Slide 5.
As Dan indicated, total sales were up 11.8% in the third quarter. On a daily sales basis, they were up 13.6%, which is an acceleration from up 10.6% in the second quarter. Mansco contributed 130 basis points to this growth, which is consistent with the prior quarter. The third quarter finished strongly with September's daily sales growth coming in at up 15.3%, or up 14% excluding the impact of Mansco. In terms of end market tone, the third quarter felt a lot like the second quarter. Macro data remained favorable with the PMI averaging a healthy 58.6 and industrial production continuing to grow at a low single-digit rate. As a result, we saw broad strength and further acceleration in our manufacturing markets, while construction sustained a mid-single-digit growth rate. From a product standpoint, we experienced acceleration in both fastener and non-fastener lines.
From a customer standpoint, national accounts accelerated again, up 17.3% in the quarter, with 71 of our top 100 accounts growing. Growth to the smaller customers also accelerated in the quarter, and it's notable that our non-national account customers grew better than 10% in September. This contributed to 64% of our branches growing in the third quarter, which is up from 62 in the second quarter and 58 in the first quarter. Based on sustained strength in most of our end markets and strong momentum in the growth drivers that Dan covered on Slide 4, we feel good about our top-line momentum entering the fourth quarter. Now to Slide 6. On a year-over-year basis, our gross margin was 49.1%, down 20 basis points versus the third quarter of 2016.
If I look at our organic fastener and non-fastener product categories, there really was very little change in the gross margin. The very driver to the annual decline was the effect of product and customer mix, the inclusion of Mansco, and the very slight drag afforded by the hurricanes. We think that these combined probably provided about 30-40 basis points of drag to the annual gross margin figure. On a sequential basis, our gross margin was down about 70 basis points. Again, the organic margin for our fastener and non-fastener categories were mostly unchanged, and the price-cost dynamic was mostly stable. Seasonality and storm effects did play a role, but the single largest cause of the sequential drop in the period was the behavior of what we call the organizational variables.
In the quarter, this includes things such as more third-party shipping, foreign exchange adjustments, lower vendor freight credits as a result of really good inventory control in the period, and other items of that sort. Which of these variables move in which direction can be hard to predict in any single period, and in most quarters, it winds up being a series of pluses and minuses that even out. In the second quarter, these skewed heavily in favor of gross margin, and in the third quarter, that favorable skewing completely reversed. Given these movements over the last two quarters, frankly, we look at the best representation of our gross margin probably being the fact that our year-to-date 2017 gross margin of 49.4% is broadly comparable to our year-to-date 2016 level of 49.5%.
Year to date, we're down about 10 basis points despite the mix effects, the acquisition, and those sort of factors that we've talked about in the past. Our operating margin was 20.2% in the third quarter, up 20 basis points year-over-year. The 20 basis point decline in gross margin was more than offset by the 40 basis point of leverage over operating expenses. Consistent with how our model works, we've leveraged operating expenses every quarter of 2017, that's resulted in our year-to-date operating margin expanding to 20.5% from 20.3% year to date in 2016. Employee-related costs were up 12.3%. 130 basis points of this increase relates to Mansco's headcount, plus the incremental expenses related to implementation of last year's DOL rules. Having one fewer selling day also had a slight negative impact.
The remaining increase is a function of the reset of incentive comp throughout the organization, given our return to strong growth in 2017, and an increase in overall staffing. Total headcount growth remained modest, at up 1.9%, or up 3% on an FTE basis, but we have added personnel in eight of nine months this year. It does remain a goal of ours to leverage the employee-related expenses over time. Occupancy-related expenses were up 1.5%. Branch costs are slightly lower, reflecting closures over the past 12 months, more than offset by growth in our Vending installed base. Selling transportation-related expenses were up 2.7%, reflecting our ability to leverage our vehicle movement activities even while we support the growth of our business. Total incremental operating margin in the third quarter was 21.3%. If you exclude Mansco, that incremental margin was 22.1%.
Frankly, we think that that value is likely reduced by more than 200 basis points by having one fewer selling day in the period. Turning to Slide seven. We generated $163 million in operating cash in the third quarter. That's a record amount for any third quarter, and it's 114% of net income. Year-to-date, we've generated $456 million, which represents 107% of net income, an improvement on the 100% in the first nine months of 2016. These improvements are primarily a function of better earnings. Net CapEx in the third quarter was $24 million, down 67%, due in large part to the absence of spending on Vending machines related to the Lease Locker program. Our anticipated capital spending target for 2017 is unchanged at $127 million. In the third quarter, we spent $92 million on dividends and $26 million to repurchase shares of stock.
Our level of debt was mostly stable at $440 million, at 17.8% of total capital, we view our balance sheet as conservatively capitalized with ample liquidity to continue to invest in our business and pay our dividend. In terms of working capital, we were particularly pleased with the inventories. If you exclude Mansco, our inventories were up 7%, which significantly trails the growth in sales. This reflected the ability of the field to leverage the heavy investment in branch inventory in 2016, and just more energy enterprise-wide on this line. Receivables growth, excluding Mansco, is up 15% in the quarter. As in the second quarter, that it grew faster than sales reflects the acceleration in growth that we saw as the quarter progressed, as well as the relative growth of national accounts. Payables, excluding Mansco, are up 23%.
Last year's third quarter was unusually low in the wake of high CSP 16 related purchases in the prior quarter, a dynamic that's likely to repeat in the fourth quarter this year. That's all we have for our formal presentation. With that, we'll turn it over for questions.
Thank you. Ladies and gentlemen, if you have a question at this time, please press star and then the one key on your touchtone telephone. If your question has been answered or you wish to remove yourself from the queue, please press the pound key. Our first question comes from the line of Robert Barry of Susquehanna. Your line is now open.
Hey, guys. Good morning.
Hi, Robert.
I just wanted to understand what's happening with price cost. I know you've talked in the past about benefits of the long supply chain, but I think the inflation would be upon you. Are you getting price to kind of fully offset the inflation?
Yeah. I'd answer that in a couple of ways. In terms of the price that we referred to in the second quarter, we felt like in the second quarter, that resulted in a positive price cost dynamic. As you roll into 3Q, we feel like we got a little bit of incremental price, but we also began to see a little bit of the incremental cost flowing through. Frankly, those probably balanced out, relative to where we were in Q2. Yeah, we feel like the price that we envisioned getting, we got.
The costs have begun to come through. We talked about that perhaps hitting in Q4, that will certainly be the case, but we saw some of that hit in Q3 as well to result in sort of a stability in that dynamic, if you will. There is still inflation. That hasn't changed. I think I would probably answer the question of how we address that the same way I've answered in the past, which is that, if at some point we determine that we can't protect our level of profitability without resorting to some sort of price action, we will take that step we believe that we would be successful in that. We didn't do anything along those lines in the third quarter.
Gotcha. Just as a follow-up, I think 4Q gross margin is typically down quarter-over-quarter, right? I know you also have a tough comp this year, should we expect that sequential pattern to hold? Maybe something in the high 48% ballpark? Is that kind of how we should calibrate 4Q?
We try not to predict, obviously. I do think you're right that we have a very difficult comp versus the gross margin that we put up last year. I think that that's certainly going to be a feature of the fourth quarter. In terms of what we expect for gross margin in the fourth quarter, there is typically seasonality. On the other hand, we did have some minor drag related to the hurricanes in this particular quarter. It's difficult for me to kind of give you a number, but we're not going to go towards the number. I think that there's some puts and takes that at the end of the day, the seasonality is usually fairly modest, and we'll see the degree to which there's some things that offset that.
Yeah. Was the hurricane about 10 basis points?
Hey, Rob?
Yeah. I just wanted to clarify what the hurricane impact was on the gross margin.
It was probably five to 10 basis points.
Got you. All right. Thank you. Sorry, Dan.
This is Dan. I'm going to chime in with a quick comment. This is the part of the call that always drives Holden probably a little crazy, but he's getting used to it. If I think about the comments he made earlier, and I think about our July call, there was quite a few questions that really were asking, what lifted gross margin? Holden's message over time has been one of, when you think about the growth drivers, the math of the growth drivers should lower our gross margin each year, optimistically 20 basis points, pessimistically 25 to 30. That's just the fact that the growth drivers, the Vending, and in the case of this year, our acquisition, tend to weigh down.
Over time, the growth of our non-fasteners relative to fasteners, which are obviously influenced by the growth drivers, weigh it down a little bit. Organizationally, what can we do to work back, to grab back some margin in our sourcing, our logistics, our selection of products, everything. In the second quarter, we stumbled over ourselves a little bit trying to explain to you all in your questions why we weren't confident that gross margin was a new normal. We talked about a lot of the puts and takes, and everything was a put. This quarter, as Holden mentioned, a lot of the puts and takes lean towards takes, not puts. Our Fastenal gross margin was down about 10 basis points. Our non-Fastenal gross margin was completely unchanged from Q2 to Q3, I think that's an important component.
It was really the organizational, the other stuff, all worked against us this quarter. History has told me, and my history perspective is 20 years, four quarters a year, so that's about 80 quarters. History has told me that you don't get things that go all your way or mostly against your way, typically in a quarter. Going into the fourth quarter, and I don't like making predictions, but going into the fourth quarter, I'm not looking for our gross margin to be below 49 from the standpoint of sequential weakening. We'll get a little bit benefit from the hurricane number. Now, in all honesty, there could be five basis points of hurricane in the fourth quarter because in Puerto Rico, we don't know all the answers. We do know that all of our employees are safe in the Gulf Coast, in Florida, and in Puerto Rico.
In Houston, we had two branches that were destroyed by the hurricane. What we know initially about Puerto Rico is we had no branches damaged. It sounds like we came through it reasonably well. Our business suffered, but our assets did not. More importantly, our employees came through it safely.
Thank you. Our next question comes from the line of Ryan Merkel of William Blair. Your line is open.
Thank you. First question is on incremental margins. Just given the 12% top line, I would have expected incremental margins at 25% plus. My question is this still an achievable target if we adjust for restoring incentive compensation, hurricanes, and one less day?
Yeah, organically, the incremental margin came in around 22%, and we believe that the day itself probably cost us the rest of the gap between that number and about 25% in the quarter. If you think about it, that extra day cost us about $18 million in revenues. Now bear in mind, of course, we know about that in advance, of course. The day's not a surprise, but it still does cost us year-over-year $18 million in revenues. We don't get a discount on our base salaries, which are a significant portion of our total employee-related expense, and there's other elements within the operating expense category that we don't get a discount on just because we have one fewer day.
We do believe that the organic incremental was around 22%, and if it weren't for that extra day not being there, we think that it would have been in that 25% range.
No change really long term to high single digit sales growth. You can get 25% or so incremental margin.
In the short term, I would agree to that. The real question, Ryan, is when you look out to the future, what percentage of our growth is coming from the Onsite model? Again, in the quarter, it was about 30% of our growth. What percent of the growth is coming from Onsite, and what does that do to weigh down that incremental margin? We're very mindful of managing the expenses around that. I don't know if Holden touched on this. Sometimes the discussion at the board meeting yesterday and discussion on the conference call get muddy for me, but I don't know if he touched on it. If I repeat, I apologize. Right now we're anticipating Q4 branch-level occupancy. By occupancy, I'm talking about rent. Utilities change with the season. I'm talking about rent expenses.
Right now we're expecting Q4 will be lower than Q1. I don't know if that's ever happened in the history of Fastenal. We're being very mindful of managing that expenses because we do believe we can get in the short term incremental margins in the mid-20s. I can't speak to four and five years from now if they're mid-20s or if they're closer to lower 20s. I know we'll be growing our returns and our cash flow handsomely at that time.
Yeah, I would just chip in two things. One, that math is basically correct. Bear in mind, again, the difficult gross margin comp in the fourth quarter, however, will make it difficult on a single quarter basis to do that. To Dan's point, we're trying to grow this business. We're doing a fairly successful job of it, given where our revenue growth rates are compared to the marketplace and things of that nature. We are going to continue to grow and outgrow the market and gain market share. Whereas we are going to manage our expenses to be able to achieve a very healthy incremental margin, and we do think about that double digit growth and that 25% type of incrementals.
Sometimes I get questions about why it can't be more, why it can't be more, and the answer is because we not only manage our expenses, but we also manage our expenses to be able to grow. I think the math that you're talking about makes some sense to us.
Thanks, guys. I'll pass it on. Thanks.
Thank you. Our next question comes from the line of Scott Graham with BMO Capital Markets. Your line is now open.
Hey, good morning to both of you.
Morning, Scott.
Morning.
Earlier this year, we were talking about how we sort of wake up in the morning with a 20 to 30 basis point gross margin drag that at that time you said you were confident in being able to backfill owing to mix and what have you. It sounded to me, and maybe I'm just reading too deeply into this, but the growth drivers now do seem to be That 20 to 30 basis points, and that does not include what would seem to be product mix, which has been running negative. I'm wondering if the growth drivers include customer mix. Is it still 20 to 30, or maybe is it now higher than that?
Yeah. A couple of things. First, I don't think we ever said we were confident that we'd backfill. I think what we said is that we are going to have a 20 to 30 basis point headwind, just as you say, waking up in the morning at the beginning of the year, and that we'll certainly work very hard to try to backfill that. Over an extended period of time, we would not be surprised if our gross margins are lower. That is on the back of significant growth and market share gains, right? I never said confident, but we certainly do wake up every day knowing we're in a hole and making an effort to try to fill it up. There is certainly never any guarantee that we would do that year after year.
With respect to the mix question, no, the mix is not different. When we talk about the 20 to 30 basis point drag, that really does fill in the product. It fills in the customer. That is really trying to take into account all the variables. The reason it does that is because I look at that mix relative to how our margins are performing across the product set, which would encompass everything that we're doing from a growth standpoint. No, I think that 20 to 30 basis points that we wake up to every year, I think that's the number. Now, we talk about how three-quarters of our revenues are fasteners and Onsites and national accounts and Vending that we believe that we have a significant moat in. Those also happen to be our growth drivers.
They're going to become a bigger part of the mix, and they also happen to be a lower gross margin. If in a year or two, that 75% becomes 80% or 85%, does that 20 to 30 become 30 to 40? Maybe. That's a function of very defensible revenue and market share gains. We'll try to fill in that hole too, but we certainly would not guarantee it.
Understood. I apologize if I misquoted you there. It does stay at 20 to 30.
Hey, Scott, hold on one second. Daniel L. Florness wants to chip in before we go to the next topic.
I'm going to just chime in. If your definition of backfill is at the gross margin line, the answer is different than if your definition of backfill is at the operating margin line or operating income line. In the last 12 months, if I look at Q3 to Q3, we gave up 20 basis points of gross margin. We picked up 20 basis points of operating income. I read that as we backfilled the 20 with 40 basis points of operating expense leverage from a backfill perspective. At the end of the day, the numbers that really matter on the income statement are further south than the gross margin line, and we backfilled that two times over. That's the key to the business.
That was done in an environment where, as you know from reading our proxy, or if you read through our proxy, you'd see that our incentive comp is really centered on paying out a piece of our profit dollar growth in the case of our leadership team, and that's true of many of our support areas, as well as our logistics costs in the case of distribution, and gross profit dollar growth or sales and gross profit dollar growth at the branch and Onsite location. In the current quarter, our incentive comp is up between 25% and 30% because we're reloading incentive comp. In that environment, we picked up 40 basis points of expense leverage. When we get through 2017, that reload is largely complete. We have a little bit of reload in the first quarter, in the second quarter, our incentive comp had dramatically expanded.
We have the benefit of that when we get into the second quarter of 2018, partially there in the first quarter of 2018.
Got you. That's great color. Thank you both. My follow-up is a little simpler. The acceleration in sales in September was impressive. Was just sort of wondering if you maybe could parse out the piece of, let's say, of sales that you recaptured from Hurricane Harvey. My guess is that that would be a smaller piece of it, and instead, I'm hoping that you can answer sort of which end markets look like they accelerated in September.
Yeah. To take the first question about Hurricane Harvey, it was fairly modest, and it was primarily in Houston that we would've regained some of what we lost in August. That regain was probably about a third. We're probably talking about half a million dollars, maybe a little more than that in terms of what we recaptured. I will tell you going forward, it's a difficult number for us to really understand. We'll do our best, it's a hard number to refine. I'm fairly comfortable at this point that we actually lost more in revenue in September from Hurricane Irma and Hurricane Maria than we gained in recovery in Houston. The hurricanes were a net negative in both August and September. That's what I would say with regards to that. In terms of end markets, it really gets down to the manufacturing end markets, right?
If you look at our heavy equipment, it accelerated. If you look at heavy manufacturing. Honestly, our manufacturing business as a whole accelerated across the board. You talk to the RVPs, and they still talk about oil and gas is still doing fairly well and all the flow-through that comes with that. The manufacturing complex in general, I think, continues to accelerate. The transportation complex for us, which doesn't include a lot of automotive, bear in mind, that continues to expand for us. There weren't a lot of areas that I could point to that were soft or weren't getting better.
That's great. Thanks a lot.
Thank you. Our next question comes from the line of Luke Junk of Baird. Your line is now open.
Good morning. First question on Onsites. I realize we're still early stage here relative to when you really put your foot down on the accelerator in terms of signings. What I was wondering is, how is the revenue run rate tracking on average relative to the numbers that you laid out at the 2015 Analyst Day? That is the about $1 million of year one market share gains and $1.8 million in average annual spend over time. The reason I ask is with the pace of signings we're seeing right now, just doing the math on those numbers alone, I would certainly speak to some really sustainable top-line momentum coming from that business.
Luke, if I look at what we're seeing, I have a few observations. One is we're continuing to see really good trends in the numbers. Averages are in line with what you're seeing. Averages can be deceiving. You have some that are well above it, some that are below it. That's the nature of averages, I guess. Very pleased with the run rates we're seeing from the standpoint of what we would've expected. One thing that's a tremendous positive for me, when I look at the layers of Onsites we're signing, we're actually seeing better performance out of the 2016 signings that we've been implementing in the latter half of 2016 and early half of 2017 than we saw in the 2015 signings that we were signing 12 months earlier. How much of that is a rising tide?
You all know, the ISM has continued to inch up. That's usually a leading indicator, so I'm optimistic that's a sign of things to come. We're definitely seeing strength in our end markets. We're also seeing underlying better performance in the Onsites we've signed, again, in 2016 versus what we did in 2015, which is really positive from the standpoint of the quality versus our historical perspective. I feel very good about that. Hope that answered your question, Luke.
The only thing I'll add to that is we continue to see the Onsite contribute more and more to our growth rate. I think this quarter, if you take out what moved from branches to Onsite, and if you remove that element of it, the Onsite contributed close to 3.5% of our growth year-over-year. That is an increase from where we were in Q2, which itself was an increase on where we were in Q1. It continues to contribute more.
A follow-up question, just as it relates to the operating leverage looking into next year. Dan, you'd already touched on this to some extent, but I was just wondering if you could put a finer point on the benefit to your growth and pre-tax income as growth in the incentive comp normalizes to get into the second quarter of next year and beyond. I think in the past, you've maybe put this in terms of how much pre-tax earnings growth that the growth incentive comp is consuming, say, right now, where we're in this adjustment period, versus a period next year where that's going to normalize to some extent.
Yeah. I'll let Dan provide some historical perspective on these numbers since he has it, given his 80 quarters I think he counted. I think we usually think about it in the first year of that reset year, if you will. It probably consumes 25%-30% of our pre-tax dollar growth can sort of go to incentive. When you get into the second year, that can perhaps get cut close to in half in terms of how much of your pre-tax it consumes.
Thank you.
I'll take Dan's silence to mean I got it right.
You got it right.
Thank you. Our next question comes from the line of Hamzah Mazari of Macquarie. Your line is now open. Again, Hamzah Mazari of Macquarie, your line is now open. Please check your mute button.
Hello? Hello, this is Kavon. I'm filling in for Hanzo. Sorry, I had the line on mute. Could you give us a sense on what you're hearing on Section 232 and the duties on semi-finished steel products, whether that's going to have an impact on margins?
We don't know yet. I think a decision on that keeps.
It might be helpful to explain to everybody else on the call what he's asking.
Yeah. This is basically a question about tariffs on Asian steel, I believe, which could encompass some of our products. Effectively, a decision on whether or not to enact those sort of tariffs continues to get punted, I think at this point, the expectation is maybe a decision comes next year, early next year. What I would tell you is I think that if the costs to procure steel products and fasteners go up because of an act like that, they will go up for everybody, Fastenal as well as our competitors. Given that it's difficult to procure a lot of product domestically at this point, the typical response would be to try to pass that through to the customers, we expect that the industry would probably take that action if something like that were to occur. We're watching it.
Nothing's been decided on it to our knowledge, if something is, we'll react appropriately.
All right. Thank you. It was quick follow-up. Could you give us a sense of how much your current business is government and how that's trending, including any exposure to state versus federal?
Our government business is probably about 4% of our total revenue. It's almost entirely state and local. We really don't do a lot with federal. By the way, that's one reason when people want to talk about sort of the recovery element after a storm, we don't necessarily have that sort of FEMA and federal government connection that perhaps some other distributors may have. Ours is primarily state and local. It's growing well for us. What I would tell you is it's probably a small enough piece of our business that we wouldn't be a proxy for the market. The last couple of quarters, we've had some nice wins. We have some energy that we're putting into the state and local government business, and we think that's providing some benefit to us. Yeah, it's about 4% of our revenue.
Okay. Thank you so much.
Yep.
Thank you. Our next question comes to the line of Matt Duncan of Stephens. Your line is now open.
Hey, good morning, guys.
Morning.
Morning.
First question I've got is just, Holden, if you look at or Dan, either one. If you look at the top 100 customers, you've got some where you're not growing and some are declining double digits. Just curious what end markets those customers are serving. Is there a common theme among them, and do you see an opportunity to maybe see growth pick up even further if some of those customers do return to growth here?
Yeah. If I look particularly at sort of where September was, within that top 100, I'll tell you, there are not many categories that are declining. The one or two that are, there's only one or two companies in each. I'm not sure if that's particularly indicative. As I said, I think that the performance, for the most part, has been pretty uniform across our end markets. Now, obviously, the largest piece of those 100 is going to be in manufacturing, and that's doing really well. There's growth across our landscape, frankly. I would say if I look at trends, I think that the Engineering and Construction, the Engineering and Construction business, If people are done with questions, but I see we are at 9:45 A.M.
The incremental margin, because we're expanding our team, we're expanding our distribution capabilities, we're expanding the expenses of the business. Generally speaking, a high piece of the gross profit dollars flow through. If I look at the expansion of incentive comp, Holden was taking a little broader brush when he was looking at his 25% number. If I look at the expansion of incentive comp at the branch, support, and leadership levels across the company, our expansion of incentive comp consumed about 11% of our incremental margin. History has shown that number's probably in a more typical year, probably not at five, but probably in that 7% to 8% neighborhood.
To Ryan Merkel's earlier question, there's probably about three points of incremental margin that are absent this year because of the expansion of incentive comp that wouldn't be there next year because it's already reloaded. Hope that clarified that point. I see we're at 9:45 A.M. Again, I want to thank everybody for participating in the call today. I hope you find this exchange useful in your understanding of the Fastenal business. Again, to the Fastenal team, nice quarter, everybody. Thanks.
Ladies and gentlemen, thank you for participating in today's conference. That does conclude today's program. You may all disconnect. Everyone, have a great day.