Good afternoon. My name is Frederica, and I will be your conference operator today. At this time, I would like to welcome everyone to the Costco Q2 earnings call and February sales conference call. All lines have been placed on mute to prevent any background noise. After the speaker's remarks, there will be a question and answer session. If you would like to ask a question during this time, simply press star, then the number one on your telephone keypad. If you would like to withdraw your question, press the pound key. Thank you, Mr. Richard Galanti, CFO. Sir, you may begin your conference.
Thank you, Frederica, and good afternoon to everyone. I'll start this tip by stating that these discussions will include forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. These statements involve risks and uncertainties that may cause several actual events, results, and/or performance to differ materially from those indicated by such statements. The risks and uncertainties include, but are not limited to, those outlined in today's call, as well as other risks identified from time to time in the company's public statements and reports filed with the SEC. Forward-looking statements speak only as of the date they are made, and we do not undertake to update these statements except as required by law. In today's press release, we had three things to discuss. We reported our second quarter and fiscal first half 2017 operating results for the 12-week and 24-week periods ended February 12th.
We also reported our monthly sales results for the four-week reporting month of February, which ended this past Sunday, February 26th. We announced our plans for a membership fee increase in the U.S. and Canada effective this coming June 1st. For the 12-week fiscal second quarter, earnings came in at $1.17 a share, or $0.07 below last year's earnings results of $1.24. Some items of note. First, our co-branded credit card and how it impacts our results. Similar to what we reported in our first quarter results, the Citi Visa co-brand credit card program, which went live last June 20th, positively impact our margins by 16 basis points, our SG&A expenses by 24 basis points as compared to a year earlier, and our overall bottom line in Q2, benefiting earnings by $0.16 a share. More detail on that later in the call. Second, gas profitability.
Our profits from gasoline during the quarter as compared to last year's second quarter were lower by $42 million pre-tax or $0.06 a share. This is primarily a function of last year's very strong second quarter gas profit results and is consistent with the impact to gas profitability that rises in gas prices will cause on our earnings of gas. Number three, IT expenses. Our IT activities impacted SG&A in Q2 on an incremental year-over-year basis by $26 million pre-tax or seven basis points, or to SG&A or $0.04 a share compared to last year. This reflects both the direct expenses for the quarter as well as increasing levels of depreciation amortization on major completed projects that are now in service. Number four, stock compensation expense. This one is getting less of a negative impact each year, so it's a little smaller than it's been.
It was 10% higher year-over-year, so $10 million higher or about $0.015 impact to the P&L. FX, number five. There are two FX items to point out. The first one, which I typically point out, is how the impact is to changes in foreign currencies relative to the U.S. dollar year-over-year. As compared to a year ago during the second quarter, many of the foreign countries and the locations where we operate began to strengthen during the quarter versus the U.S. dollar, most notably in Canada, resulting in our foreign earnings in Q2 when converted into U.S. dollars being slightly higher by about $4 million pre-tax or about $0.01 a share than if the exchange rates had been flat year-over-year.
Number two, as it relates to FX, is a much bigger impact to this quarter's P&L, had to do with FX losses related to forward contracts and U.S. dollar holdings by our international subsidiaries. These are used to pay U.S. dollar-denominated merchandise payables in those countries. That those losses on those exceeded the gains on the related U.S. dollar-denominated payables. This year in the second quarter, it was roughly a $20 million pre-tax hit. Last year in Q2, the gains on those payables exceeded the net losses on forward contracts and U.S. dollar holdings by +$6 million. Year-over-year, a $26 million year-over-year swing or negative impact to the P&L by $0.04 a share. I might add that this year-over-year swing, it generally runs in the ±$0.00-$0.02 a share range.
With all the volatility out there, it was a little bigger this quarter. Number six, LIFO. There was no LIFO charge or credit in this year's second quarter results, whereas in last year's Q2 results, it had a LIFO credit of $15 million, reflecting deflation in our LIFO indices, and so positively impacted last year's Q2 by $0.02 a share. While we did have some deflation in the quarter, as we did in Q1, with the switchover to a new accounting system and platform at the beginning of the fiscal year basically even though we've had deflation, we have no associated previous inflation or LIFO charges historically taken. If you will, there's a buildup of credit that will offset future LIFO charges to the extent that there's inflation in the future. Again, year-over-year, that's a $0.02 hit to the quarter year-over-year swing.
Number seven, income taxes. Last year in Q2, our effective income tax rate was right at 34%. Due to a discrete tax item this year in Q2, as well as small changes in the profitability mix by country, our effective rate this year came in rather at, instead of the 34%, came in at 35.6%, effectively impacting our Q2 EPS by about $0.03 a share. Turning to our second quarter sales. Reported sales were up 6%, and our 12-week reported comparable sales figure came in at up 3%. For the quarter, the 3% plus comparable sales figure was helped by gasoline price inflation to the tune of about 84 basis points. While the impact from FX was a very slight detriment, again, while currencies strengthened during the quarter, the net over the quarter was still a slight detriment of about -9 basis points.
together, about three quarters of a percent hit. Excluding gas price inflation, the reported +3% U.S. comp remained at 3%. Our reported Canadian comp of +8 was actually a +2, excluding gas inflation and FX. Remind you that the Canadian dollar strengthened quite a bit. The reported -2 other international comp, excluding gas and FX, would've been a +3. In total comps reported for the quarter at +3% for the quarter, excluding gas and FX, with the pluses and the minus still remained at +3%, ex gas and FX. For our four-week month of February, which included the last two weeks of the fiscal Q2, the first two weeks of February did, comps came in at a +4 on a reported basis.
This consisted of a +5 reported in Canada, a +10 reported in-- I'm sorry, +5 reported in the U.S., a +10 reported in Canada, a -2 reported for other international. As I discussed last month, or actually as we discussed on our monthly sales call last month, the calendar shift of the Chinese Lunar New Year, that was 11 days earlier versus last year. That positively benefited the January reporting period this year and negatively impacted February. We estimated this shift was a detriment to February comps of about three quarters of percent on the total company and 6.5 percentage points, so 650 basis points to the other international segment.
Sales in February were positively impacted by both gasoline inflation to the tune of a little over 200 basis points and by overall strengthening in foreign currencies relative to the dollar to the tune of about +60 basis points. Ex gas inflation in the U.S., the reported +5 would've been a +2. Ex-gas inflation and FX in Canada, the reported +10 would've been also a +2, the -2 reported for other international would've been a -1 ex gas and FX and a +5 excluding the Lunar New Year shift. Total company comps for the month reported at +4 would've been a +2 excluding gas deflation and FX. I'll also point out cannibalization. We do that every quarter and typically it's somewhere in the half percent range or a little less.
Cannibalization has become a little bigger of a factor to our comp sales results in the last couple of months. The cannibalization impact on February was approximately 90 basis points to the negative for the total company, and it was actually -75 in January. For February, it was -300 basis points in Canada. Mind you, this year we're opening, I believe, seven new warehouses on a base of 91 up there, a lot of relative cannibalization for Canada. -180 basis points on the other international segment in February. I mentioned that the -90 basis points in February for the total company, by comparison, for all of fiscal years 2015 and 2016, total company cannibalization averaged a little under 40 basis points to the negative. Again, it's picked up late of late with some of the openings.
We estimate weather had a negative impact on February comps as we had snow in the east and heavy rains in the west. The estimated impact in that was about 50 basis points to the U.S., about 75 in Canada, and to the total company, so about 50. Regarding deflation overall and primarily in the U.S., we've seen deflation in the 1%-1.5% range in February. Departments such as foods, sundries, frozen foods, liquor, meat, deli showed the most deflation on the foods and sundry side. On the non-food side, consumer electronics continued to be deflationary primarily in the TV category. In terms of new openings, our opening activities, we opened a net of eight new locations during the first quarter, nine less one relo. In Q2, we opened four new locations.
Those included our 13th unit in each of Korea and Taiwan, as well as two new locations in Florida, in the Tampa, Florida area. For all of fiscal 2017, we have current plans of 29 net new locations. 17 additional openings during the third and fourth quarters of fiscal 2017 are planned. Of the 29 for the year, 14 in the U.S., eight in Canada. I'm sorry, eight in Canada. I mentioned seven earlier. It's actually eight on a base of 91 in Canada. One each in Japan, Korea, Taiwan, Mexico, and Australia, as well as our first openings in France and Iceland, most likely in mid to late May.
This afternoon, I'll also review with you membership trends and renewal rates, our membership fee plans in terms of increases in June, an update on the Citi Visa Anywhere Card, an update on our Multi-Vendor Mailer promotional activities, additional discussion, of course, on margins and SG&A, and a little bit about e-commerce results and some initiatives there as well. In terms of second quarter results, quickly on the sales. For the quarter, sales were up 6% to $29.13 billion. On a reported comp basis, they were up 3%, and again, ex gas and FX, they still remained up 3%. For the quarter, the +3 reported comp was a combination of an average transaction increase of a little over 1% and an average shopping frequency increase of 2% for the quarter. That's company-wide. The frequency in the quarter for the U.S. was a three.
In terms of sales by geographic regions, the Midwest, Texas, and Northwest regions were strongest, with California not far behind. Internationally, in local currencies, better performing countries were Mexico, U.K., and Korea. In terms of merchandise category sales for the quarter, for the second quarter within food and sundries, overall flattish year-over-year. Liquor or spirits and foods were the leaders. Tobacco continues to be a negative and actually in the high teens, as we've mentioned, that we look to cycle the majority of those tobacco sales losses by the end of June. For hardlines, overall in the low-to-mid single digits. The strongest department results were tires, hardware, and seasonal, with consumer electronics down in the low singles. Softlines were also up in the low-to-mid single-digit range, with apparel and home furnishings showing the strongest results.
In fresh foods, comp sales were also in the low to mid-single digits. Lastly, in the second quarter overall, again, in terms of deflation, for the second quarter was in the 1.5%-2% range. Similar departments on the foods and sundries side. Non-foods, again, saw a little deflation in consumer electronics, primarily TVs. For February, traffic was up approximately 2% and 3% in the U.S., including 3% in the U.S., while average transaction was up a little under 2.5%. Most of this was due to gas inflation and FX. In terms of geography for February, Midwest, Texas, San Diego region, which also in our case includes Arizona and Colorado, and New Mexico, were the strongest, as well as the Bay Area. Internationally, in local currencies, U.K., Mexico, and Canada were at the top of the list.
From a merchandise category standpoint, excluding FX, food and sundries and hardlines, including the consumer electronics, were up low single digits, softlines up mid-single, and fresh foods slightly negative for the February reporting period. Again, a little deflation is impacting these numbers. Before moving to the income statement, a few comments about our Multi-Vendor Mailer, the coupon booklets that we send out and have online, what we call the MVMs, these promotional activities, and a few changes we've recently implemented. As most of you know from many years ago, the MVMs have grown and evolved over 22 years from one six-week summer coupon booklet back in 1995 to generally year-round promotional price pieces with great values on items being offered in each mailer. Over the years, we've expanded the mailers and have continued to tweak them.
More recently, we've revamped the MVM program, creating some newness, enhancing member values on many of the items, and creating a little bit more merchandising excitement. We've eliminated a few of the MVMs over the course of the year, also there'll be fewer days. We've reduced the number of items per mailer, but we've overall increased the offering in terms of total savings of those items. We're also moving, in some cases, to everyday low pricing, EDLP, where we can drive higher over sales and show better everyday pricing and value. In terms of Q2 2017, it was really the transition fiscal quarter for these MVM changes, if you will. The first revamped MVM ran in December, which is near the beginning of Q2. Overall, in the second quarter, we had 17 fewer MVM promotional days.
Mind you, the quarter itself is 12 weeks times seven is 84. Couple of days closed for the holidays, but basically 17 out of those 84. 71 fewer items offered in those MVMs. Again, higher overall sales compared to last year in the MVMs. Overall, so far, we like what we see. We continue to tweak it a little, but remember, we're still in the early stages of this. We know that 17 fewer MVM days in our 84-day second quarter and 10 fewer MVM days in our 28-day month of February probably hurt traffic a little. We shouldn't see that latter aspect in Q3, as there are the same number of MVM days year-over-year in the third quarter. Again, Q2 is really the transition of that. Now moving on to the line items in the income statement.
In terms of membership fees, coming in at $636 million, up 5% or $33 million versus last year, down a basis point. Minimal impact from FX because, again, while they were increasing over the course of the year, they started off lower in the beginning of the second quarter. In terms of membership, we continue to enjoy strong renewal rates, coming in a little over 90% in the U.S. and Canada and 87.7% worldwide on a fully captured basis. We continue to see increasing penetration of the Executive Membership program in the countries where we offer it. In terms of number of members at Q2 end, at Q2 end, primary Gold Star came in at 37.5 million, up from 12 weeks earlier at 37.1 million. Primary business, 7.4, up from 7.3 12 weeks earlier. Business add-on, 3.4 million, down from 3.5.
That has to do with some of those people converting into their own membership, generally as they become Executive Members. Total accounts, 48.3 million compared to 47.9 fiscal quarter earlier. Total cardholders, 88.1 million at second quarter end, up from 87.3 million 12 weeks earlier. As of Q2 end, our paid Executive Membership stood at 17.9 million, which is an increase of about 200,000 from 12 weeks earlier or about 17,000 additional per week. Executive Members represented a little over a third of our member base and about two-thirds of our sales. In terms of renewal rates, again, overall in the U.S. and Canada, 90.2, down a tenth from 90.3 at the end of the quarter, which was also 90.3 at the end of the fiscal year back in late August.
Business within that remained at 94.3 in both fiscal one and two quarters end, and primary Gold Star at 89.5. Again, it probably a little roundy, just push it down instead of up that tenth of a percentage point. Worldwide actually picked up a little at Q2 end. It was 87.7, up from 87.5% at Q2 end, and 87.6 at fiscal year end. We feel these are pretty good numbers, and don't really see a lot of impact. We believe a lot of it has to do with the conversions in credit cards. If you recall me talking probably a year and a half ago, about some of that stuff we saw in Canada as we transitioned the Canadian credit card program a year and a half or so earlier than we did in last June here in the U.S.
If I look back, as an example, just a year ago in Q3, Canadian renewal rate was a 90.6. For this quarter, it was a 91.6, so it started to come back as we would expect. Again, U.S., we're still seeing some of that auto bill impact that we believe is a big piece of it. Back in Q3 a year ago, we were 90.3, and actually 90.4 the prior quarter, down to 90.1 at the end of the year, and we're at 89.9 at Q2 2017 end. Again, pretty much the kind of impact that we would've expected to see and not really terribly concerned about that at all. I want to spend a minute regarding our announcement on the increases of fees this morning, which will be effective June 1st. First, the planned increases relate to our U.S. and Canadian operations.
Recall that membership fee increases took place in several other countries effective this past September 1st at the beginning of the fiscal year. In both the U.S. and Canada, which by the way represents just under 90% of our company's fees, about 87% or 88%, the current annual fee for our individual Gold Star business and business add-on memberships, what we refer to as our primary memberships, is currently $55 a year, and has been at that level since November of 2011, about five and a half years ago. The annual fees for these memberships will go to $60 effective June 1st. Also, in the U.S. and Canada, our $110 per year executive membership fee, which has been at that level also since November of 2011, is being increased by $10 to $120. Also with regard to executive membership, the 2% reward associated with the executive membership will increase.
Currently, the annual reward is capped at $750. That will be increased to $1,000. While there is an increase in the annual fee, the reward goes up to $1,000, and that's based on eligible purchases. That, of course, is in addition to the 2% reward that one gets if they use the Citi Visa Anywhere Card at Costco, or the 4% when they buy gas at Costco. In all, approximately 35 million member households will be impacted by this increase, approximately half of whom are executive members and half of who are primary Gold Star business and business add-on members.
Note that the membership fees are accounted for on a deferred basis, so in terms of how it hits our P&L, our membership income line, approximately one twelfth, if you will, of one month worth of the increase in fees from the June renewers, that'll be the first group that gets this fee increase. Approximately one twelfth of the increase will be booked to the income statement in that first month of June, with an additional one twelfth being booked in each of the succeeding 11 months. Next, increased fees from our July renewals. Those will be booked starting in July one twelfth and following through to the following June and so on.
The full P&L impact of these increases will be over a 23-month timeline, such that the last group of members to be billed at these new levels will be next May of 2018 with a booking, if you will, of those $5 and $10 increases being recorded over that month and the succeeding 11 months, i.e. 23 months out. Before continuing down the income statement line items, a quick update and a few updated stats on the Citi Visa Card offering. Again, this began last June 20th, early in our fourth quarter of 2016. Recall that we began last June 20th with approximately 11.4 million co-branded cards, or about 7.4 million accounts being transferred to Citi for conversion to the new Citi Visa Anywhere Card. As of Q2 end, just under 90% of the accounts transferred have been activated.
Recognizing all accounts transferred to begin with were not activated, I think it was down in the low 80s% at the time, low to mid 80s%. In fact, that just under 90% activated as of Q2 end, that's up a few percentage points from Q1 end 12 weeks earlier. Also, we now have about 1.2 million new approved member accounts representing about 1.6 million new Citi Visa cards out there since the June 20th conversion. This is also up about 200,000 accounts over the past 12 weeks during fiscal second quarter. Lastly, we are seeing the Citi Visa co-brand portfolio total spend higher year-over-year, both organically from the cards converted last June, as well as from these new accounts. We'll see what the next few quarters bring.
Overall, in terms of conversion, usage, and new sign-ups for the card, we feel it's going pretty well so far. Going down to the gross margin line, our gross margin in the second quarter on a reported basis, was lower year-over-year by 24 basis points, coming in at 11.00% compared to last year's 11.24%. As usual, there's a lot of moving parts here. Gas inflation and the impact of the credit card, some of that benefit goes to the sales line, which therefore improves the reported margin. I'll ask you to do my little matrix here, we'll do it for first and second quarters. There'll be 4 columns. Reported Q1 2017. Without gas deflation Q1 2017 is the second column. Third column is reported Q2 2017, the last column would be without gas inflation in Q2 2017.
The first line item would be core merchandise. In Q1, reported +19 basis points year-over-year, without gas deflation, +16 basis points. For Q2, reported +1 basis point, and without gas inflation, +9 basis points. Ancillary businesses, -5 and -6 in the 2 Q1 columns. In Q2 2017, in the 2 columns, -20 basis points and -18 basis points, again, reflecting lower margins in gas year-over-year, while increasing the penetration of gas sales. 2% reward, -2 basis points and -1 in columns 1 and 2, and in columns 3 and 4, 0 and -1. LIFO, -2 and -2 in Q1, and -5 and -5 in Q2, again, recognizing that a year ago we had deflation and therefore LIFO credits.
This year, while we have deflation, we can't take them, since there's nothing to take them against from prior offsetting LIFO charges. Other, last year, there was, I believe, a one-time legal settlement that added 19 basis points to the first 2 columns here, and 0 and 0 in columns 3 and 4. You add all that up, last year in the first quarter, year-over-year margins were up 29 basis points on a reported basis and up 26 basis points ex gas deflation. I also mentioned last year that within those numbers, the Citi Visa impact to margins within that 29 and 26 was +13. In the next 2 columns, the reported Q2 2017, again, margins on a reported basis came in 24 basis points lower year-over-year. Ex gas and FX came in 15 basis points lower than last year.
Mind you, both of those numbers do include the benefit from the Citi Visa program to the tune of about 16 basis points to the positive. Again, taking those out, the reported -24, adding the 16, that would be a -40, and adding 16 to the -15, it would be a -31 on an adjusted basis, if you will, using Citi Visa. The core merchandise component of gross margin was higher by 1, as you see in the chart, and +9 excluding gas. Excluding the benefits of Citi Visa, it was -15 and -7 excluding gas inflation. As I always mention, subcategories within the margin are core subcategories, food and sundries, hardlines, softlines, and fresh foods. As a % of their own sales were positive year-over-year by seven basis points.
With the declining sales penetration of that, given the inflation in gas, the contribution is a -7. Food and sundries and hardlines were both slightly higher year-over-year on their own sales. Softlines was up about 60 basis points, and fresh foods was lower year-over-year by about 10 basis points. Ancillary and other business gross margin, I mentioned, was down 20 in the quarter. Most of the year-over-year decrease was due to lower gas profits, as I mentioned earlier in the call. 2% reward, one basis point of negative impact to ex-gas, just implying a slightly higher sales penetration, ex-gas inflation on a year-over-year basis. I already mentioned LIFO. That was about $0.02 a share as well.
Overall, our margins, ex the Citi Visa credit card benefit, were most negatively impacted by lower gas margins, somewhat negatively impacted by five basis points from LIFO year-over-year, with slightly lower year-over-year sales penetration of core also hurting it a little bit, even though core margins on the core sales were up seven basis points. I might also mention that the +7 basis points core margin improvement year-over-year, this is notwithstanding some of the pricing initiatives that I mentioned earlier, like EDLP, that we've been taking these last couple of months. Moving to reported SG&A, our SG&A percentage in second quarter year-over-year was lower or better by five basis points, but higher or worse by 4 ex-gas inflation, coming in at a 10.23% this year compared to 10.28% last year on a reported basis.
Again, excluding the benefits of the Citi Visa program, which clearly helped SG&A by lowering merchant charges. Excluding benefits from that, year-over-year, SG&A was higher by 19 basis points and 28 basis points ex-gas inflation. In terms of the performance year-over-year in SG&A, operations component, as I mentioned, was lower or better by eight basis points year-over-year and +1 excluding the impact of gas. You see that in the chart we just drew. The +8 basis point improvement consisted, again, of much lower Citi Visa merchant fees and related fees, somewhat offset primarily by higher payroll and employee benefits costs. Again, that has to do with the underlying sales being a little lower and a few other things. Central expense was higher year-over-year in Q2 by two basis points, 3 without gas.
IT was seven or eight without gas of that and offset by a couple of things that went our other way. Stock compensation expense, again, one to two basis points, not a big amount. Moving down to pre-opening expense. Pre-opening expense was higher by $5 million, coming in at $15 million in Q2 2017 versus $10 million a year earlier. Pre-opening relates not only to the actual openings net quarter, but also some of the ones leading up to it or getting ready to be open, rather. That was four openings in Q2 this year, only one opening last year. There's also higher year-over-year pre-opening expenses related to our entrance into two new countries, France and Iceland, as we already have people on the ground. Operating income in Q2, all told, came in at $844 million, 1% lower or $12 million lower from last year's $856 million.
Below operating income line, reported interest expense came in at $31 million in both this and last year's fiscal second quarters. Interest expense, I might mention that interest expense at beginning partway through Q3, I believe the middle of March, will improve quite a bit with a scheduled March 15th, $1.1 billion debt repayment. This is a 10-year fixed rate debt instrument, I believe about 5.5% fixed rate interest. Net will save about $50 million a year pre-tax, about $60 million of reduced interest expense on that pre-tax, offset by cutting a check for the $1.1 billion, if you will, and losing some interest income on that to the tune of roughly $10 million, and that's just a simple guesstimate. Interest income and other was lower year-over-year by $20 million in the quarter. I was just informed that I skipped the SG&A chart.
Why don't we go back and write that just so it's easier for those of you that put it down. Thank you. Four columns for SG&A, reported and without gas deflation, Q1 2017 and Q1 2017, and reported and without gas inflation, Q2 2017 and Q2 2017. I think these numbers that I just read will make a little more sense. Operations, Q1 2017 reported minus eight basis points or higher by eight basis points, and minus six without gas deflation, plus eight and plus one in columns three and four. Central, minus nine and minus nine, and Q2, those columns three and four, minus two and minus three. Stock compensation, minus seven and minus six, and minus one and minus two in Q2.
Mind you, it's always higher in Q1 because we do our big total company, total employee, those that get our issues grant in October. Other, plus eight and plus eight last year. That was a year-over-year unusual item, I believe, and zero and zero in columns three and four. You add it all up on a reported basis in Q1 2017, it was higher year-over-year on a reported basis by 16 basis points and without gas inflation by 13. On a reported basis in Q2 2017, we were lower or better by five and higher or slightly worse by four basis points without gas inflation. As I mentioned, the Citi Visa impact to SG&A in Q1 was 25 basis points. That's in those numbers, and 24 basis points in Q2. I won't go through the numbers I just mentioned.
My apology for doing that in reverse order. Below the operating income line, reported interest expense came in, I mentioned, at $31 million, and I also mentioned that come March 15th, we're going to save about $0.08 a share or $50 million pre-tax net, $60 million reduced interest expense, and roughly about $10 million pre-tax reduced interest income. I was just getting ready to talk about the next line item on the income statement, the interest income and other. It was lower year-over-year by $20 million in the quarter. This year in the quarter, it was a -$4 million number. Needless to say, it's the other, not the interest income. Last year, it was +$16 million. Actual interest income and other than the FX that I talked about earlier for the quarter, were better year-over-year by $6 million.
Offsetting this, of course, was that $26 million in charges related to various FX items that I discussed at the beginning of the call. Overall, pre-tax income was lower by 4% or $32 million in the quarter, coming in at $809 million versus $841 million a year ago. I mentioned earlier in the call, income taxes, a little higher rate, coming in at 35.6% this fiscal quarter versus 34% last year. We think for the year, our current best guess for effective rate for the year is actually a little lower than that 35.6% that we recorded in Q2, probably somewhere more likely in the 35.3% or 35.4%. That's our best guess. Overall reported net income came in at $515 million compared to $546 million net income last year in the same quarter. A quick rundown of other topics. The balance sheet is included in this afternoon's press release.
A couple of balance sheet items that we try to give out. Depreciation and amortization for Q2 total $312 million for the quarter and $609 million for the first half of this fiscal year. In terms of AP ratio, accounts payable as a percent of inventories, 92% reported both this year and last year. If you take out non-merchandise payables for construction and other things, this year it came in at 82% AP ratio, and last year was 83%, just rounding down a little there. Average inventory per warehouse came in at $13.1 million compared to $12.8 million a year ago. We're up about $350,000. Variances, about a little over $100,000 of it's health and beauty aids and the like, over-the-counter items. Some of that has to do with the timing of our MVM buildup in February.
Others are mostly non-foods, majors, about $100,000, small electrics, about $75,000, and hardware, about $50,000 to $60,000. So nothing terribly out of pocket there. In terms of CapEx, first quarter, we spent $670 million. I'm sorry, in the second quarter, we spent $515 million. In the first quarter, we spent $670 million. Total year to date, $1,185,000,000. We estimate for the year, CapEx will still be in the range of $2.7 billion-$2.8 billion, so a billion and a half to go, if you will. A billion and a half plus to go. Next, Costco Online. We're in the U.S., Canada, U.K., Mexico, Korea, and Taiwan. Korea and Taiwan being the most recent additions about a year ago. For Q2, sales and profits were up, of course. Total online sales were up 12% and up 11% on a comp basis.
As I'd spoken about a little bit, probably in Q4 last year in October, and then a fiscal quarter ago back in early December, we continue to improve our offerings, merchandise offerings, and enhance our member experience online. In terms of improving merchandise, we continue to add new and exciting merchandise, and we continue to improve in-stocks on high velocity items. In terms of recent initiatives and recent online additions include Samsung appliances and a variety of added apparel brands, both direct and indirectly purchased, with additional offerings in various non-food categories in the coming months. These would include Kohler bath and kitchen, Reebok men's and women's active wear and footwear, and Spyder ski and outerwear, to name a few.
We've also continued to add various health and beauty aids, both regular ones and upscale ones, and sundries items to our online offerings, increasing both online page viewing and member shopping frequency as well. One recent fun item to note leading up to Valentine's Day, we offered online 50 long-stem roses for $49.99, and that included delivery. We had sales in those three days of over $2 million, represented over two cargo planes, and most important, that $49.99 price point for 50 long-stem roses is $30 less than what we sold the same item for just a year ago. If you're late to getting those roses, we are now offering them at $39.99, including shipping.
In terms of improving experience and functionality of the site, we've improved search, we've shortened the checkout process, and we've improved our members' ability to track their orders, and we'll continue to do some more of that. Just recently, we automated much of our returns process, not only providing members much better quality of service, but also reducing by more than 20% in just the first couple of months our call center volume related to returns. There'll be more to come over this coming calendar year in several of these things, and we're finally getting around to doing some of this. Lastly, we continue to improve our distribution logistics. For example, in the U.S. and Canada, we now fulfill online orders from 11 depot distribution points, which of course allows for closer and faster delivery of online orders.
Overall, as I mentioned, we have plenty of online initiatives going on, both in terms of member experience and service, expanded products and services offerings, and greater value to the member. Finally, a quick update on other home and office delivery sales channels. As you know, we partner with Google Express in five cities, operating out of 15 of our Costco U.S. warehouses. In addition, we're working with Google Express on a new service offering 1 to 3-day shipping of products throughout the continental U.S. We also continue to work with Instacart. Instacart currently operates in 26 U.S. cities, in our case, utilizing 132 of our U.S. locations. We're either testing or getting ready to test two other third-party delivery services within the next month or so. In terms of expansion, in fiscal 2016, last year, we opened 29 net units, about 4.5% square footage growth.
This year, it looks like it's going to be also 29 net new units. It's about 4.25% square footage growth. The planned fiscal 2017 locations, these 29 locations by country would be 14 in the U.S., eight in Canada, and one each in Taiwan, Korea, Japan, Australia, Mexico, and then the new ones in France and Iceland. As of Q2 end, total warehouse square footage stood at 105.1 million square feet. Next, in terms of stock buybacks, as you recall, in Q1, we repurchased $122 million, or 809,000 shares of Costco stock for an average price of right around $151 a share. In Q2, we repurchased $66 million or 411,000 shares at an average price of just under $160, $159.86, I believe. In terms of dividends, our current quarterly dividend continues to stand at $0.45 a share.
On a quarterly basis, this $1.80 per share annualized represents a total cost to the company of right at $800 million. Lastly, before I turn it back to Frederica, our fiscal 2017 third quarter scheduled earnings release date for the 12-week period, third quarter ending May 7th, we'll do that after the market close on Thursday, May 25th, with the earnings call that afternoon at 2:00 P.M. again. With that, I'll be happy to open it up to Q&A, and Frederica, I'll turn it back over to you for that process.
At this time, I would like to remind everyone, in order to ask a question, please press star, then the number one on your telephone keypad. That's star one to ask a question. Your first question comes from the line of Simeon Gutman with Morgan Stanley.
Hey, guys. It's Simeon. Richard, I think the big question that I think will come up a lot, it looks like, I'm not in front of all the numbers, but margins in this quarter sequentially versus Q1 seem like they got a little bit weaker. I'm excluding, I think, some of the gasoline impact, which granted, I don't think was fully captured. Is anything changing as far as investment back in the business, SG&A dollars, e-commerce, or just reinvesting back into price sort of that we're not seeing?
Well, certainly the gas inflation does have an impact on it, as well as does the increasing sales penetration of gas. That's probably as much of an impact. Certainly, some of the more promotional stuff we did with everyday low pricing had an impact. I didn't bother to try to quantify it because it's lots of different items, and that's what we do. If anything, we did it a little more because we wanted to. There's been some other things going on. Nothing major, though. I look at the numbers, and I don't see a big change other than things I mentioned and perhaps what I just mentioned about that. Oh, yeah. As Bob is reminding me, and again, getting back to the gasoline comment, a lot of it has to do with the contribution penetration of these areas.
With gas inflation, I think gas prices were up 29% year-over-year, and gallon comps were up higher as well because of that, were higher because of what we do. We have good prices. That had probably a bigger impact. Gas is typically, these are rough numbers, 800 or 900 basis points lower than the rest of the merchandise that we sell. It could be 700 or 1,000, but it's big, and you have increasing penetration of that. Not only were gas margins down with rising gas prices within the gasoline business, but the impact that it has-- and again, if you get back to what I mentioned on the core margins on their own sales, which is roughly what, 80% of our sales, food and sundries, hard lines, soft lines, and fresh foods.
On core margins year-over-year in Q2, they were up seven basis points, but the contribution, if you will, to the total here was minus seven.
Got it. Okay, that's helpful. Then I guess my follow-up, just thinking about reinvestment rates in the business, right? You have a membership fee increase coming down the pike at some point, and now we know when, and there's a lot of debate on how much flow through or not. I guess maybe the way to ask it is there a natural run rate of reinvestment that you as managers of this business try to put back in and sort of where are we running relative to that? Yeah, I'm trying to gauge, when we get the membership fee increase, are we going to see that get pushed back into price as well, or is that some of that's going to come through to the bottom line?
My view, I'm speaking historically here, that roughly we've done increases about every five years, we're certainly not smart enough to figure out how to put it in over the five years. We know it hits the membership fee income line over about two years. We also know that it hits very little in the first month, one twelfth of the renewals, if you will. Historically, we've invested a lot of things, including continue to invest in price, which might even be a slight negative then. Arguably, we've chosen with some of the revamping of the MVM and some of the things we're doing with everyday low pricing to, if you will, do some of that now. That really is, in my view, not directly related to if when and now when, not if, a fee increase.
I would argue that we probably did a little more now, certainly the reduced number of every day days. Even in the MVM, what we've done is we've basically gone out to vendors and worked with them. The goal for us and them is to drive sales. Basically, it's a better value to the member, which means it's a little more expensive both for the vendor and us. I would say in some ways we started that process, or we always do that process, and you'll see when you see. Generally speaking, in my view, it historically goes in over a longer period of time, not just those two years, the net of everything.
Okay. Thanks, Richard. Good luck.
Your next question comes from the line of Michael Lasser with UBS.
Good evening. Thanks a lot for taking my question. Richard, presumably you're investing in price because you're seeing a more competitive environment. Can you quantify or at least qualitatively comment on how much more promotionally and competitively intense it is now than when you've seen it in the past, particularly around the time when you've raised your fees?
Well, you know what, we have a couple of senior merchants in the room with me shaking their head because I concur. This is really not terribly related to that, to the fact that there's increased levels of competition. When we look at our direct competitors, notably direct warehouse clubs and certainly supermarkets on certain key fresh food and sundries items. A lot of what you read about were some of the big box discounters and their investment in price, which is formidable. Our view is that impacts, and the competitors have to deal with that more directly, are supermarkets, in case of use a Walmart or Target or whatever else. We really haven't seen a big change there. Certainly this was not motivated by that.
I think more of it has to do with is over the last few years, you see some of the sales lift and some of the things that the MVM has been great for us for so many years changing. It's an iterative process. If anything, just like we are ultimately, we're going to do the right thing for the long term, irrespective of how it impacts now. We knew. We chose. We don't give direction. We knew that this was going to have some impact, not a big impact, frankly, but with having less days by going to some EDLP. It's what it does to drive sales. I don't know where else you can get in the country a 40 pack of half liter water bottles for $2.99 down from $3.49. We're driving units. We're driving a little traffic.
That's what we do. It's not because somebody else went down to that price. We look at some key items and how do we do this. I think it has very little to do with either a change in the level of competition or the fact that we're getting ready to do a fee increase. We do that really those somewhat independently.
Just to quantify the impact of pricing, you got a 16 basis point benefit to your gross margin from the credit card transition. It sounds like you invested maybe all of that and more back in the everyday low price.
I think the same comment. We didn't say, "Hey, we got this much back. Let's use it all or let's use any of it." We really look at those independently. I think one of the comments I made on the Q1 call, which is the first quarter that there was really some transparency of how big of an impact and successful the new credit card has been, both in terms of adding to our gross margin percentage and reducing our SG&A percentage in terms of lower fees. We really didn't say, "Wow, that's so big, let's use it." We do what we're going to do on pricing, and this is where the chips fell on that.
I think one of the comments made on the first quarter conference call is, we were more lucky than smart that that helped offset some of the reductions and things, this is what we do.
Okay. My follow-up question is, the assumption is because you tend to skew higher on the socioeconomic income demographic or your membership base does, that you were not impacted by the delayed tax refunds in February. Do you think that's right, or do you think there was some impact, and if so, could you quantify it?
Well, I can't quantify it, certainly. Historically, when we've been asked similar questions about when there was issues with food stamp programs or things like that, or this, we generally don't feel that we're impacted a lot at all. It certainly is not a positive, but my guess is it's not very much of a negative either. I would lean towards saying not really a big impact at all.
Okay. Thank you.
Your next question comes from the line of Christopher Horvers with JPMorgan Chase. Christopher, your line is open. If your phone is on mute, please unmute.
We can move on to the next
Your next question comes from the line of John Heinbockel.
Richard, when you think about the Executive increase in particular, how much discussion was there, what you should do with that in light of where your club competitors are, where Prime is? When you think about where that goes over time, what kinds of things can you add? I mean, obviously, the 2% is a big benefit. What can you add to the Executive that people might want that could allow that to continue to go up over time?
Again, first of all, what we do best is value to the member on products and services, and certainly that's going to be part of it. As it relates to enhancing both the primary membership and more importantly, the Executive membership, I will say, and I'm not being cute, stay tuned. We got other things that we're planning to do.
Yeah.
Nothing's earth-shattering, but there's several things we'll do, including, I'm sure, one late summer, early fall. We'll continue to do that, and we'll continue to look at the value proposition on the credit card. As I mentioned, I'm sure a year ago when it first transitioned to it, whatever we do, and certainly the credit card profitability and success has been better than I think we and our partners had originally planned. We'll look to enhance that value over time, but that's not something you do until you're a couple of years out to see where things trend out. We'll continue to be very good at driving value and getting you to say, "Wow.
Again, sort of when you think about what you could have done with Executive, was there even any serious consideration, leave it alone, enhance the value proposition, and push more people toward Executive? You pretty much knew you wanted to do the $10 from the beginning.
Well, I think we always start with the premise that we pretty much knew that. We look at other things as well, we'll see what happens in the future. Nothing's going to happen in the next year or so, I'm sure. There's some other reasons in the state of California that it makes sense as it relates to sales taxability.
Yeah.
We're pretty simple, it works for us. We're pretty simple, we're pretty extreme in terms of value. Yeah, we always talk about, what about another level of membership above that? Do you ever figure out how to just offer only an Executive? We're not there yet. Rest assured, there'll be more value-oriented things coming to the Executive for sure over time, more value to all members in general pricing and things.
All right. Then just one last thing. When you look at KS assortment over time here. Clearly it's grown, looks like maybe up 50% over the past 5 years or so. That's imperfect, right? Because there's stuff that cycles in and out. When you think about where that ends up 3 years from now, 5 years from now, obviously your total assortment is staying somewhat the same, are we going to get a higher percentage of KS items just continuously almost forever?
Sure. I think yes. Our success has been based first by selling branded goods at the best value out there. It's for two reasons. One, such sharp savings relative to everybody else.
Yeah.
Two, our KS, if it's as good as, if not better quality, which is our starting point, even a greater savings versus what we sell the brand for, that's even better. It keeps our members happy, it keeps us and our vendors honest, and we'll continue to drive it. Recognizing there aren't a lot of three and $500 million items out there like the waters and the paper goods and the K-Cups and whatever, but there's lots of $20 million and $30 million and $50 million items that we get surprised every day. So yes, there'll be a continued push for that, but there's also a continued push to add brands that historically haven't been prepared to sell us.
Okay. Thank you.
Your next question comes from the line of Matt Fassler with Goldman Sachs.
Hello, Richard, and good afternoon. Couple questions. First of all, just to make sure that we understand the magnitude of the impact of the change in the MVM and the move to EDLP. When you talk about EDLP, are you solely talking about stepping in and displacing the MVM? Is there a way to estimate the impact that that might have on margin? Does that impact grow or dissipate next quarter as the day disparity for the MVM comes down?
Yes and no to each of those partial parts of those questions. We've tried some items using the MVM example. Some we've chosen to, with vendors, chosen to go to EDLP, and get out of the item, get out of the MVM. Others, we've gone the other way. Sometimes it works one way, and sometimes it works the other way, and that's how you figure that out. Again, not to be smug, we really don't see it as, yes, small basis points impact $25, $29 billion quarterly sales figures by a lot. We knew it was going to impact us this quarter. We don't provide direction, so we really can't say anything. It should improve a little in the next quarter. That's not based on, "Oh, my God, let's change it." It's just that's what happens.
Is the improvement based on the better sales that you're likely to get from the MVMs year-over-year or some other element related to the change you're making?
It's all the above, including, as we've only done this for a couple of MVMs here. A subset of all the vendors over the years that have participated in this with us, some that participate with us all the time for 22 years, some that are in and out and seasonal, some that are new. As we see the things that work, we certainly don't keep that a secret, and we go to our vendors, and we have vendors in some examples that are wanting to do more for us, wanting to get those prices even lower because they see the lift impact on it. It really is all of the above.
I joke and use the word a little strategery because we are merchants, and we try a lot of different things, and we're pretty good at figuring out what works and what doesn't and working with our vendors to do that.
One quick follow-up. Where are we right now in the credit card benefit cycle for you? Are there elements associated with Citigroup Visa, whether it's sign-ups, whether it's outside sales and the financial benefits they have to you that would lead to the financial benefit to you to ramp up, or are we kind of leveling off at what's likely to be sustainable?
Well, I think there was some benefit in Q4, but not big enough to actually separate out. The big kahuna was in Q1 when we reported it, now to Q2. Again, very simply, my guess is sometime you'll get some incremental benefit in Q4 to get the other half of it, if you will. It's not exactly half of that. In addition, though, there's probably a little more in Q4 because of some of the challenges of the conversion itself. Then beyond that, I think I've said occasionally that it should and hopefully will be the gift that keeps on giving a little as we drive more penetration both in Costco with the 2% component reward on that, along with the 2% executive member reward.
It'll drive outside spend because the fact that the card is accepted at so many places, and if we can get you to use it everywhere or most everywhere, particularly those small merchants that generally pay higher fees to everyone, there's a component of that that we benefit from. There's no way to predict what's going to happen. I think on a year-over-year comparison, the biggest bang for this buck is over the first 12, maybe first 15 months because of the first 8 or 10 weeks had the challenges of the conversion. Beyond that, there should be incremental benefits, but certainly not as big as the first year. And by the way.
Thank you so much
by the way, at some point, two or three years out, I'm guessing you're going to see, when we get comfortable, see where it goes, you're going to see, again, enhanced improvement because that's what we do on even the reward proposition.
Thank you.
Both ours and our cards.
Thanks.
Your next question comes from the line of Paul Trussell with Deutsche Bank.
Hi, this is Tiffany Kanaga on for Paul. Thanks for taking our questions. I know you've gone through all the numbers in quite some detail, but can you help us understand a little better as we're lapping quite a significant deleverage in core SG&A excluding gas. We didn't see the leverage this quarter that we'd hoped for. Can you walk us through what might have held back the core SG&A line ex gas this quarter and how to think about it going forward as we continue to lap even greater deleverage?
Well, at the end of the day, it's mostly, as it always is, payroll and benefits. Healthcare is inflationary.
We still have bottom of scale increases.
We still have bottom of scale increases that we've done, and that was actually in March of last year. That's when Natal anniversary, I think mid-March. That hurt us a little bit. Again, we look at the numbers. I think we always look in the mirror at each monthly budget meeting and say, could we have done a little better on controlling labor in the warehouses, controlling overtime hours? Sure. There's probably a few basis points that we could do better on. For the most part, I think it's the underlying comp. Mind you, using that, again, silly example of the water, when we go from $3.49 to $2.99 and sell a lot more units. The margin dollars are plus or minus the same, but there's more labor involved.
Not a lot more, again, all these things are incremental, add a few basis points. Other than that, the big one, of course, is all the modernization-related stuff. As we've had these big systems come on, it's like building a building. You build it, then you wait until you open it, if you will, or turn it on, and that's when you start to amortize it. In our case, typically over five and seven years, sometimes three on a few things. Some of those big nuts, if you will, have occurred in the last year.
Thank you so much.
Your next question comes from the line of Brian Nagel with Oppenheimer.
Hi, good afternoon.
Hi.
I, too, wanted just to drill a little further into this, into the margin. Just to be clear, and I know you've discussed this a lot already, Richard, but if I'm looking at my model, year-on-year gross margins were down, ex membership fees, down by 24 basis points. That's the first time in a while, at least in the several quarters now, that it's been down. Did this shift from the MVM to more EDLPs is a significant portion of that? I'm sorry, ask the last part of the question. Somebody's saying something here. The question I asked, and I know this is a bit repetitive because others have asked similar questions, but if I look at the gross margins, ex membership fees here in the fiscal Q2 were down 24 basis points.
Right.
That's the first time, as I look back sequentially from that same map, that they were down. If we look at that break in trend, how much of that can be explained by this shift from the MVM to EDLP?
I would say if there were five factors, I'm making this up, it's probably the third or fourth most impactful, and the biggest single factor, which is well more than half of the factors in total, is gasoline. Gasoline impacts you two ways. One, you have substantially lower margin on gasoline sales. Two, you have increasing penetration of gas because of a 29% higher price per gallon. This is a business, I don't know the exact number in my head, but it's at roughly 10% of our total company sales. It's ranged from nine to 12 or 13. That alone, again, if I go back to Q2, if we look again on a reported basis, lower by 24 basis points, ex gas inflation -15, but that -15 includes -18 of ancillary, most of which is gas.
That's where a lot of it is.
Okay. On the shift from MVM to EDLP, is it fair to assume that if this is implemented, it could have more of a near-term impact on margins, then maybe it takes a while for sales to catch up? I'm thinking there is it, because the MVM to a certain extent is a call to action to consumers, correct? If you shift to an EDLP, maybe you lose that call to action over time, the consumer figures it out.
Well, a couple of things here. First of all, it's not like we've changed this thing. It's not binary. We've not gone from all this way to all another way. It is the transition quarter, and we did a bunch of stuff in that transition quarter. Again, that's, in my view, one of the smallest factors relating to this decline in margins year-over-year. The comment I made earlier to one of the questions on the call about what do we see going forward, I think Matt had asked it, is that it should not be an issue. Yes, we'll see, but we don't believe it'll be an issue in Q3 like it was in Q2. I really wouldn't lose a lot of sleep at this juncture over that, and certainly not lose any sleep over the fact that we've made this major change.
The major change is we've changed it. We're seeing the things and components of it that work and seeing some of the things that we're tweaking. I think one of the reports out there that I saw a couple of days ago, it talked about adding some days into the thing. No doubt, have some friends in the vendor business. The comment I made earlier about Q3, whereas in Q2, on an 84-day quarter, there were 17 less MVM days. In the four-week month of February, 28 days, there were 10 or 11 less MVM days. In Q3, in that 84-day month, less a day or two for holidays, there are the same number of MVM days year-over-year. Again, I can't overemphasize that. EDLP is not a word we use a lot historically. We're trying a lot of different things.
We feel pretty good about what we've got going forward here.
Got it. Thank you.
Your next question comes from the line of Oliver Chen with Cowen and Company.
Hi, Richard. Thanks for the details on the roses, too. I like that. Regarding the mailer and the changes historically, what were you seeing with the consumer in terms of the consumer insights that drove you to engage and test reacting on the changes to the MVM program? What are you looking for as it gets tested? As we model the membership fee increases, how do you think that will interplay with store traffic? Do you feel like store traffic will continue in the nicely positive range, or will there be some volatility and risk we should think about going forward?
First of all, as it relates to how and when we decide to change some MVM stuff, not that we talk about this all the time, but for those of you who've known us, we've talked for the last few years about how we tweak it a little bit. Ultimately over time, there are items that don't get, not only the same annual sales lift that they had historically gotten, but also the additional incremental new potential customers to that given item from the vendor's perspective. It's not just getting somebody to buy more of something, to consume more of something, it's getting new members to do it. Over time, some of those things, not all of them, some of them get stale. We've done some tweaking over the last few years.
Probably six months ago, three months ago, the merchants sat down with the operators and senior people there and made a choice to do what we're doing now. We knew and we know it was the most painful in that transition quarter. Again, in our view, again, not to be cavalier about it's not that big of a deal other than it did impact some things this way. Again, we continue to tweak it. Unfortunately, because small basis points make changes on EPS numbers and we sell at a nice multiple, I recognize the concern, but we feel we've got a lot of good things going on, and we're pretty optimistic about what's going to go on here in the upcoming quarters as it relates to driving sales and improving earnings.
Again, I can't give you specifics about how and why, we're pretty also good, as you know, laying out all the stuff here, good and bad, you can take a look at it. We don't look at this and see it as that big of a deal, we have tweaked it a little bit. We have added a few more days back to some things. We knew also that, again, Q3 was going to be a lot less of any of those negative impacts as it relates specifically to MVM days.
On the membership-
I want to get back to the EDLP question because, again, it's not an acronym in our vernacular, even though that's what we do since the beginning of time. When we look at these items and we work with different vendors on different items, sometimes it's let's put a little more emphasis on the MVM, change the pack size, do a greater value. Sometimes it's let's get out of the MVM and do an EDLP and really come to the table with something that is a wow on an ongoing basis. Again, not just the last couple of months, but over the last few years, as we've tried this occasionally, we've seen different things. Again, we feel good about what we feel that we're going to see going forward.
We feel good about the lifts that we've seen in many of the things, but we are by no means going to change everything to go to EDLP, which used to be MVM, by any stretch. We think the MVM has a lot of continued potential for us. By the way, there's some added benefits to it. Historically, in the last few years, our MVMs, I believe there was three days between the MVMs in the warehouse. When you're talking about lots, what are the things that take up the most room and require the most operating issues in the warehouse? It's big bulk items like detergents and paper goods and water and the like. Logistically, we pushed that pretty hard over the years to three days. One, this gives them a little bit more efficiency in the warehouse.
That's not the reason we did it, but that's an added benefit. Two, we've actually put some time in there to do a few other things in the warehouse, whether it's road shows or some other things we've got going on with some merchandise presentations. All of this relates only to what we do every day. We keep trying new stuff. It's not like, oh my God, something's changed.
Okay, Richard. Thank you. What about your digital infrastructure in terms of, is curb pickup and buy online, reserve in store, as well as your mobile, as well as thinking about the bricks and clicks supply chain, what factors will be important over the next few years in terms of making sure you're driving ease, value, convenience, as well as being superior merchants?
Well, first of all, in terms of pick and click, no. Ask me next quarter. I think the answer will be no for a while. You ought to go see it in some of the other places, by the way. It ain't all that good. We don't have the room for it anyway. We are doing some unique things with business delivery, and we now have whatever, 15 or 16 business delivery sites around the U.S. getting ready or just opening our first one in Canada. I think we just opened our first one in Canada. Probably the single biggest thing is what items we're offering online and how quickly we can get them to you. I mentioned, maybe it was a little bit of a teaser, but I talked about online.
You'll see some more things coming this calendar year, probably not until summer, that both what we're doing ourselves as well as what we're doing with a couple of third parties, not only in our markets, but outside of our markets in the continental United States. Look, one of the things that we all know, we have the best prices on great quality items, and we've never been too good about worrying about how to get it to that end customer a day earlier. It's the 80/20 rule with us. Just in the last six, eight months, we've done a lot of improvement online in that customer experience with the smallest amount of effort, the low-hanging fruit. We've got some good things working on, but we're doing these things honestly from an offensive standpoint, not a defensive standpoint. I'm not trying to be cute there.
Clearly, we want to do it for competitive reasons, too. It's not like we looked at this and we've lost. We see our renewal rates, X some of that auto bill stuff that we believe. We see our traffic going up still. Online, we see our page views and the like going up as well with some of the additional items and the types of items we put on, and better communication to our members of what that is. I think, again, sometimes we're viewed as the tortoise, not the hare. Certainly, over time, we're viewed as being stubborn. I think in my view, we're a lot less stubborn, but we're still a little bit of a tortoise sometimes, and we got a lot of good things going on. We'll see. Stay tuned.
Our last question, Richard, is on stores. A big topic is just physical stores. You obviously have really superior traffic and a great assortment. What is your maximum store opportunity in the United States versus where you are currently? Are there any potential edits to the format, or is it status quo in terms of what we should think about over the longer term?
Well, I think, first of all, that every year it seems to occur that we think there becomes a few more locations than we thought were possible. Certainly, in the last three fiscal years, in 2015, I think we opened 21 in the U.S., as an example, and 16 or 17 last year, and I think 17 I said this year. If you'd asked me five, seven years ago, I'd guess we'd be down to 10 to 12 a year. Clearly, we've gone into several new markets, new for us, not new for the warehouse club industry, where we've done well. Some of them are smaller and take a little longer, but we're doing just fine there. I think the line will keep getting a little bit more towards that saturation, but a little slower than some might think. We don't know.
In terms of our basic 150,000, 160,000 square foot format Costco warehouse in the U.S., I think it's a good guess to assume that it'll be somewhere in the mid-teens for the next 10 years per year. I don't know. Three years from now, I might be wrong. At this juncture, we're doing a little higher than that, which is a surprise to me. Canada, this year is an anomaly, opening eight on a base of 91. I can remember five, eight years ago, I don't remember how many we had, maybe we had 70. We felt one day the market potential for all of Canada could be 90. Now it's probably in the 110, 120 range. Not seven or eight a year. This is unusual, but it's good for us. The opportunities present itself.
I don't see us necessarily anytime soon doing a unit half the size. We tried that a while back. Not to say we won't try something. We do now have, I think we began this year with 14 business centers all in the U.S. I believe we plan to end this year with 18, one of which is our first one in Canada. Could you have 40 or 50 of those in the U.S. one day? We'll see. We had eight for the first 10 or 15 years before we opened a ninth a few years back. I don't know if the 18 or the 17 in the U.S. by fiscal year-end will be 25 one day or 50 one day, but it's working, and we'll see where it goes. The second, the last leg-
Thank you.
Yeah. The lastly leg is international.
Thank you very much, Richard.
Your next question comes from the line of Scott Mushkin from Wolfe Research.
Hi, this is Ben Shim in for Scott. Thanks for taking my question. Just a couple of questions. Can you give us an idea at this point what your expectations are for inflation or deflation across some of your consumables categories, such as food and personal care products?
I think the collective view is inflationary or less deflationary for the next few months, and then maybe a little inflationary. It's a crapshoot.
Okay.
It's based on asking our buyers in different departments. There's nothing widespread at this juncture.
Okay. Going to your historical experience with respect to membership fee increases, has 10% been the norm over the last several increases as far as attrition goes?
I'm sorry, 10%? What was the 10% related to?
I think you mentioned something like a 90% renewal rate over the last membership fee increase. I'm just wondering.
Yeah, no, our renewal rate over the last many years has tweaked up from the high 80s to the low 90s in the U.S. and Canada. There was a period of time there for probably two or three years that, let's say three years over 12 quarters, it seemed like every quarter, year-over-year, was up another tenth. For a lot of reasons. Executive members, as we convert them, they renew at a higher rate. Visa, on the co-branded credit card, there's the benefit of auto billing for those that opt into that. You'll get a little higher rate there. Hopefully we keep doing things to make you want to renew more.
Okay. Relative to five years ago when you had your last fee increase, can you describe what the competitive environment was back then, and if you remember, and how it might be different now? Does that give you pause or concern?
Absolutely no concern.
Okay. All right. Thank you very much.
Sure.
Your next question comes from Robbie Ohmes from Bank of America .
Hey, Richard. I just quickly was curious on an update of your business in China, what you're seeing on Tmall, and if you're any closer to maybe opening a store over there. Thanks.
We continue to do Tmall. I think there was an issue with taxes that negatively impacted not only us, but anybody selling, importing into China. There was a 10% increase. There was a 10% increase in taxes. Boy, maybe somebody else shouldn't do that. There's a 10% increase in taxes, and that impacted, again, negatively some of the imports into there. It's still relatively small. It's good, but it's relatively small. In terms of us opening a location over there, you should expect something in the next couple of years.
Got it. Thanks.
Your next question comes from Karen Short from Barclays.
Hi. Thanks for taking my question. I just want to go to this fee increase, and I'm asking this because I'm wondering if you could give a little color on where your average ticket is with your executive members now, say, versus 2012. Obviously you can calculate what the breakeven spending is off of this new fee. When I try to back into how many trips per week your executive member needs to be making, you're almost up to once per week for that member to break even. I may be wrong on the math on the average ticket, so just wondering if you could give a little color on that.
We don't give out those numbers, but directionally, I think one of your assumptions is the average executive member spends a lot more per visit as well.
Okay.
A business member. Relatively speaking, more of our business members are Executive Members as a percent of total business members versus Gold Star. I'm guessing. I don't have the numbers in front of me. All those things play. I want to also add a comment to a question a few questions ago about renewal rates, about how they've changed over time. I have a simple summary sheet here that shows what our renewal rates were, and I only have U.S. and Canada combined, which is, again, more than 80% of our company. At the end of 2005, it was 85.9%. At 2010, it was 87.7%. At 2015, it was 90.6%. At the end of 2016, it was 90.3%.
Again, that little delta downward has to do, in our view, with the conversion to the new Citigroup Visa card and some of the changes with automatic auto billing had to be redone. Sequentially, it's continued to go in that direction. Recognizing it can't go above 100%, and jokes aside, it won't get that close, but we think that it's been consistently improving at a level that is consistently good for us.
Okay. Just on e-commerce, wondering if you could maybe give a little color on e-commerce growth rate by categories. Obviously your growth rate overall trails some of the mass competitors. Wondering, A, did it come in line with where you expected for the holidays? Or what you think you could continue to do to drive that strength?
I've shared, I think, a little bit of the kinds of things we're doing, from member experience to faster delivery to expanding items. Arguably, letting our members know that we have it. Again, it started off years ago as limited big-ticket items, hard to carry, hard to deliver, hard to install items. We've added to that. We feel, again, fine with where we are. Yes, there are bigger increases out there, but I would bet that the investment per dollar of increase is dwarfed everywhere else. I'm not trying to be cute about that because we are investing more in it, but doing it in an offensive rather than a defensive way. We're still a brick-and-mortar entity, and we want to get you in the store because you're going to buy more in the warehouse.
You're going to buy more when that happens, we've got a lot of reasons for you to do that. We also recognize that we don't want to lose a sale to somebody else because they only buy online. I feel you're going to see some good things continue over the next few years. Online is a small percentage of our company, 4%, $4.5 billion or so. It's still a $4.5 billion business growing at 11% or 12% this past fiscal quarter, and the mid-teens over the last few years in general. I'm betting that'll go up some from that level, but we have to see.
Great. Thanks.
Your next question comes from Sean Naughton from Piper Jaffray.
Good afternoon. Just wanted to go back to the merchandise margin. The core merchandise margin up 7 basis points in the quarter just on a rate basis. This has been going on I think for a couple of years now. Is it really driven by mix, or is it some of the fewer promotions you were describing or better deals from your vendors? Just trying to understand how this line item has just been a very consistent march higher, and is that intentional by Costco?
First of all, it is intentional by us. My guess is that most of it's related to mix. This quarter, a little bit's related to the EDLP and the MVM shift or whatever, but generally speaking, and certainly over the years, increasing penetration of gas. You've got our total company gross margin last year of what, approaching 12%? If roughly 10% of that's gas at-
He's talking core on core.
Oh, core to core.
It was up a little better in Q1 than Q2.
I'm sorry. I think that's just the nature of the draw that quarter. Fresh foods tends to be a little higher margin, and I think it was down a little bit this quarter. Some of that has to do with holding prices on some deflationary items. A couple of months ago, I think there was a bad berry crop. Literally, these are the kind of things that'll impact us when we're such a big player in this stuff.
Yeah.
I don't see any big difference there.
Okay. I was going say, because it just seems to me, and I'd have to look at the model again, this has been a number that's been very, very consistently going up between 5-15 basis points almost every quarter for the last probably 10 quarters.
Well, no, on core on core is not mix related, it's core on core.
I'm just saying, like within the core, yeah.
Yeah. Okay. I'm sorry. I misunderstood part of the question. I think part of that is some of the aspects of the components of that. Some of the higher margin categories would be some of the non-food categories like apparel. For those of you who've known us for a while, I think the last three years, we've enjoyed probably an annual compound growth rate in our apparel sales of certainly the high singles, maybe 10. That's a category that's a higher margin to start with. I think we've done a little better job on some other areas of non-foods. Apparel is the one that stands out in my mind as a big example. Fresh foods overall has a higher margin, and even though it fluctuates up and down core on core on its own core, if you will-
Yep
it's a higher average margin department. Increasing penetration overall.
Okay, got it. Just on new sign-ups and renewal rates, I know millennials are becoming a bigger piece of the store, and I think Craig mentioned something like 44% of new sign-ups are actually millennials now. Can you talk about how the millennials are doing in terms of their renewal rates, just overall? I guess as an add-on there is, any color on the renewal rates for some of the promos that you've experienced or experimented with over the last 2 years on LivingSocial or in social media, how the renewal rates from those programs have gone?
I can't tell you how the age group called millennials today and how they renew and spend versus 5 years ago, whatever they were called, or 10 years ago, whatever they, Gen Y or Z or whatever it was back then. We didn't look at that kind of data back then. We have for the last 2 years. Ask me in 3 years, I'll have some good information for you on that. What we see, though, in terms of that age group that are now called millennials, it's not that different relative to the other age groups today than it was 2, 3, and 4 years ago, when they sign up. What we see is, we only have I guess 2 good data points on the 2 LivingSocial things we did, about 2 and a half years ago and about a little over 1 year ago.
We compared them to everybody else that signed up that month by just walking in or going online to sign up. What we found is through the LivingSocial, I could be off a few percentage points here, that on LivingSocial, it was about in the mid 40% range of those that signed up on the LivingSocial promotional effort that were millennials. That compared to the walk-ins, that was in the mid 30s, maybe 9 or 10 percentage points difference. What we saw in terms of how much they spent over the course of the year, the LivingSocial or the millennials, spent a little less each time and actually shopped a little more frequently, which is counterintuitive to me.
In terms of renewal, they renewed about a percentage point or 2 higher than the walk-ins in that first year that they had to renew. Again, a little counterintuitive to me. Maybe it's not statistically meaningful because, again, it's the first year, it's 1 or 2 percentage points. At least it gave me comfort personally that we're not losing them, and they are coming in. What we've also seen and what we believe when I look at kind of the curve of who spends the most at Costco, they start spending more. The peak is, if I do 2, 4, 6, 8. If I separate people from 25 to 80 years old in 11 age groups, the peak is the fifth and sixth age groups, which are 45 to 49 and 50 to 54.
A nice increase going from 35 and 39 up to 40 and 44 before that. Well, maybe that makes sense. Maybe they're getting married, maybe they're having kids, maybe they're getting married a little later, having 0.1 less kids, who knows? Once they do that, and then they start making a little more money, they spend more. They have more mouths to feed, and they are making more. Again, this is looking at a chart, not doing a lot of statistically significant analysis.
Okay, that's helpful. Last quick question from me. When should we expect the golf ball back in stock?
When was this the golf ball back in stock? Stay tuned. If and when we have it back, we'll let you know.
Thank you.
Your next question comes from the line of Chuck Cerankosky from Northcoast Research.
Good afternoon, guys. I've got a question because what you said about the MVM is counterintuitive. You're very focused, Richard, always on growing frequency of visit, and it sounds like by reducing the days, there's the op. How do you communicate the other factors you don't advertise to the member about road shows and new excitement or lower prices in certain key items you're trying to feature?
Okay, well, first of all, how we do it's the Costco Connection, our magazine, email, and in-store sites. I got to tell you, we started this doing the hurt first, if you will. This goes back over a year ago, when you start talking to vendors and our key merchandise partners and figuring out what we're going to do and how we're going to do it. It wasn't like we need to get every reduced day filled with something else. Giving a little breathing room to the warehouses has been a big positive from our operators' standpoint. Again, we'll see how that goes. I don't think it was a big surprise.
We knew we would get impacted a little bit on the traffic side, and we know, I think one of the comments I mentioned, and one of the comments that some of you guys have mentioned or heard through others, is that there have been a few extra days added to the starting point. Well, that's correct. If we look back out in the next three, six, and nine months, the big extreme transition was this quarter and in particular February, and you won't see that in Q3. It's the same number of days. We'll stay more fluid in the next two to three quarters. Then it's just anniversary stuff.
All right. Thank you.
Your next question comes from the line of Kelly Bania with BMO Capital.
Hi, thanks for fitting me in. Just wanted to ask about the decision to raise the dollar impact for the 2% reward. I was just curious if you tested that increase to $1,000. If you have any thoughts on what kind of impact you think that could have on traffic or ticket from that Executive Member. Also, I guess, associated with that, how we think about the impact to gross margin as that flows through over the next several years. Thank you.
First of all, there's a lot of Executive Members. In theory, even assuming the new fee structure of $60 and $120, that $60 means that you got to spend, what, $3,000 more a year to be breakeven on it, on eligible purchases. That's not a big hurdle, but there's plenty of Executive Members that don't get near the $750. There's some that it won't be an impact at all. There's others that'll be a nice impact. How we came to $2,000 is not unlike how we decided to do five and 10 each five years or six years or whatever, and how we went from $500 to $750 last time. I think, again, you'll see some other things that we'll add to the benefit. We continue to have added things to the Executive benefit. Our family bought a Yukon Denali a couple of years ago.
If you were a regular member, on top of getting incredible pricing on the car, a regular primary earner, we got, I think, $200 or $300 cash card. An executive member got a $600 or $700 cash card. Needless say, that incentive people to become an executive member. Once they are, they look at what other rewards. There's plenty of things out there, and we'll continue to add to that. How it impacts margin? The extra $250, it's relatively small, very small relative to the 35 million people paying $5 and $10 each. Mind you, some of those are CAD. That's a little lower relative to the US dollar. Nonetheless, it's a very small piece of it.
Thanks.
Your next question comes from the line of Peter Benedict with Robert W. Baird.
Hey, Richard. Quickly, when you pay down the debt later this month, your leverage ratios drops pretty dramatically. Just what are your latest thoughts around leverage? If you care to comment on the circus in Washington and what's going on there, is that impacting your decision on when you might do something around leverage? Thank you.
Well, clearly, as it relates to all the proposed tax things, one of which is the ability to not write off interest expense, that is not one of the ones we're worrying about. We don't have a lot of interest expense. As it relates to our balance sheet and leverage, we like to think of ourselves as well capitalized, not overcapitalized. We're cognizant of it. I'm, again, not trying to be cute. We look at all the components of it, regular dividend growth, special dividend stuff, stock buybacks, first and foremost, ramping up expansion. We're cognizant of it, but again, we're not going to just do something because this is March 15th and we got to do something. Again, stay tuned. Our board meeting meets regularly every quarter. We just had a board meeting a few weeks ago and didn't do anything different.
We're constantly asked questions about, are you going to increase the regular dividend? Are you going to do another special? Stay tuned. We look at it every quarter and we decide what we want to do. At this juncture, we basically use cash to pay off that debt, and I'm happy it's the most expensive piece of debt that we have.
All right. Sounds good. Thanks, Richard.
Your next question comes from the line of Edward Kelly with Credit Suisse.
Hi, guys. Thanks for taking my question. Richard, just looking at gross profit dollar growth, I know there's been a lot of talk about margins, but gross profit dollar growth this quarter slowed relative to where you were last quarter. Is that primarily fuel? Asking in another way, did you make less in profit per gallon in fuel this year than less in gross profit dollars in fuel this year than what you did last year?
I'm guessing we did, because when prices go up, we make less per gallon, period.
Yeah.
Again, I don't have the exact numbers in front of me, but let's face it, $0.06 a share? $42 million pre-tax in the quarter year-over-year, that ain't all gallons. It's mostly profit per gallon.
Yeah.
Vast majority of it is profit per gallon .
Okay. As we think about, obviously the Street wasn't really modeling the quarter the way that it came out. There's a lot of talk about the gross margin, it really seems like it may be around this fuel side, with a bit of price investment on top of that. I guess, is that fair? My question for you beyond that is, as we look out into Q3, as long as fuel prices are stable, does this headwind go away a bit?
Bob, what did we do last year in Q3 on gas? Was it still big? I think it won't be all things being equal in terms of the comparison year-over-year, Q1 and Q2 profitability in gas a year ago was outsized big. We were profitable in Q1 and Q2 this year, but again, the comparison was a huge difference. Some of that outsized big became less outsized big in Q3. It was less in Q3, but we don't know what Q3 this year brings. My guess, it certainly won't last. There's no certainty to anything in life. My guess is it'll be a little less negative, all things being equal.
Thank you.
I'm sorry.
Your next question comes from the line of Greg Melich with Evercore ISI.
Hi, thanks. I had a quick follow-up to that one, then another question. If you're thinking about penny profit, Richard, are we back in gasoline to where we were a few years ago, or could there still be room to go down there? My other questions were on deflation. I think you mentioned in February it was about 50 basis points less than in the second quarter. I'd love to know what drove that. Maybe a longer-term question, if you just look at membership growth numbers, we have been running close to 7%, and it's decelerated to about 5.5% the last few quarters. Could you help us understand why that is, and if it's just timing the clubs or what's at work there? Thanks.
Okay, Greg, I hope you wrote down the numbers because a few people were whispering to me as you were talking. What was the first question?
First question was a follow-up on the gas profits.
Oh, yeah.
Oh.
We've had about, I don't know if it's a year and a half or two years of outsized gas profits. Is it back to where it was before two years of outsized gas profits? I believe so. I don't know if it's a little worse or a little better than two and a half years ago, but certainly these last two years have been fun.
Got it. The other questions were, deflation in February looked like it was not quite as bad. It was like 50 basis points less deflation than in the second quarter.
Yeah. I think, again, that's year-over-year, so part of that is just when the relative timing of inflation was a year earlier in those respective quarters. Probably a little of it is some lower pricing on some stuff. Again, I use the water example. That item's a $300 million item, and we did more sales, but
It was a little bit up.
What?
It was a little better than quarter, worse than January.
Bob is mentioning in terms of deflation, it was a little better quarter in Q2 versus Q1, February was actually a little more deflationary than the whole quarter, relatively speaking, than the whole quarter two overall.
Okay.
It fluctuate up and down. I think, again, some of that has to do with pricing on our side. It's not just what the economists are telling us.
All right. Then lastly on the membership growth, I know it can move around sort of quarter to quarter, but if you look a few years ago, it seemed to run at sort of seven-ish, give or take, and now it's more like five and a half the last couple of quarters. Is that because we have more infills, or how should we think about less membership growth for the same sort of club openings?
Some of it's related to the cannibalization I mentioned earlier. When you open 8 units in Canada this year, there aren't a lot of new markets. We've had openings in the U.S. where in a small new market, well, Tulsa was my extreme example, where through opening day, we had 22,000 sign up, 20,000 plus sign-ups. I remember in Tennessee, in a new market, we had 10 or 12, which is great. We could open a new unit in L.A. and have 3,000 sign-ups, and it's an awesome location because it's existing members shopping a lot more frequently because it's a lot closer unit to them, but they've always been members.
The other thing that'll affect that is international openings, particularly in Asia, where we could have during those 8 or 10 weeks up through opening day where we do tabling activities, we've had openings of 20 to 50,000, 20 to 40,000 sign-ups in those few weeks, those several weeks. Having a few of those change, and again, I don't know if that should help us or hurt us right now, just those are the things that are generally impacted. Overall, we feel that we're still adding members. Some of the new markets that we've gone into, Tulsa, again, was an extreme one, but again, we tend to do well in those.
Okay, great. Thanks.
Your next question comes from the line of Molly Smith with Bloomberg.
Hey. Molly Smith here from Bloomberg News. Thanks for taking my question. Richard, I wanted to ask you about the prospect of the border adjustment tax, given that your cost of goods sold, about half of that comes from imports. What have your thoughts been as these discussions continue? Have you tried to lobby with any of these other retailers against the tax and as well in other changes potentially coming out of the new administration with healthcare, how that may impact your business as well, if at all?
Sure. Before I answer that, just one final comment, response to Greg's question about the increased growth in new members. Actually, not members, but new members by revenue. A little of that probably has been negatively impacted by the auto billing comment that I've made about the credit card transition. A little bit, I don't know how much, but I know that's probably a little bit of that offset, too. As it relates to the border adjustment tax, there's clearly the people out there that want it, manufacturers that export a bunch of stuff and don't import a lot of stuff, and at the very other extreme, retailers. Recognizing border adjustment tax is just one element of one version of the tax reform plan that's been put forward out there.
The probability of what's going to happen and when it's going to happen and how much of it's going to happen, we don't know. We don't believe it's good for consumers. It's going to raise prices. Ultimately, I've read articles where some retailers, particularly apparel retailers, where 90+% of their merchandise is sourced overseas. Well, a 20% tax is a 20% tax, no matter how much. While retailers generally tend to historically be full corporate taxpayers, us in the mid-30s, in the U.S. probably a little higher than that total company effective rate, it's going to hit it. If it were to go through, we personally don't buy into the fact that it'll be offset by a big rising dollar. We don't know what's going to happen with the retaliation out there by other countries, and we'll see.
As a retailer, we definitely think it's bad, and we're against it. In terms of lobbying, we're not big on lobbying. We're doing a little bit in that area. Certainly, RILA, Retail Industry Association, is very involved in it. There's an offshoot organization formed, which RILA is certainly a big part of, and we've become part of the Americans for Affordable Products. We've joined that along with hundreds of retailers, including some very large retailers. Through those lobbying efforts, we certainly support what they're saying, and hopefully, those out there that will make these decisions are listening. We spent our whole lives driving down prices and recognizing also that so many items, it's not a question of let's buy them here instead of outside the U.S. They don't exist here, and that's not going to happen overnight.
It will be a tax, ultimately, in our view, and prices will ultimately have to rise.
Thanks so much. Anything on the Affordable Care Act either, or no?
I'm sorry?
Anything else there on prospects of change to ACA, if that impacts your healthcare at all?
It really doesn't. We have a very good, quality, rich medical, dental, vision and other plan, where our employees only pay about 10% of the total cost. It's one of the things that sometimes it impacts our P&L a little bit, but it's something that we're very proud of. So we really don't have a lot to say about that.
Thanks so much.
Your next question comes from the line of Joe Feldman with Telsey Advisory Group.
Yeah. Hi, guys. Joe Feldman from Telsey Advisory Group, as you know. Sorry to prolong the call, but just wanted to ask, could you share any thoughts on the service element of your offering? I'm thinking about travel and maybe, I don't know, payroll tax or doing different things, auto sales, and maybe how data mining could play into that?
All those services, we try to not talk about them a lot because we've got a lot of good things going on. They're profitable, they're growing. Many of you have heard me talk to you about, challenge you to go, next time you rent a car, go to costco.com, and no matter how smart you think you are, and you'll see what a great value it is. We're doing better at that and getting that word out, and you'll see additional services. As it relates to data mining, we're starting to take some baby steps in that area. Again, our first and foremost is we're pretty good at getting on the phone and calling third-party people that we think that can be a good partner to us, and we'll continue looking at other things.
Those are all things that'll continue, I think, to drive our business in a positive way. I don't want to suggest we're hiring somebody to do big data mining at this point. We're doing more data analytics than we've ever done, there's plenty of low-hanging fruit to start.
Thanks. One other question, just on IT expenses. I know it's been a drag much of this year, presumably it never goes away given where we're headed with technology in general. Is that the right way? Should it annualize at some point and level off, should we think about incremental expense going forward?
About three and a half years ago, when we embarked on this dark journey, recognizing we probably had the lowest cost IT out there, I always joke we were the greatest MASH unit. It was always up and running, it was band-aided to death. We've made a big investment. We also, during the process, found out what we don't know and what we need to do. Again, it's gone up. I think the best guess four years ago was incrementally it might cost in the low double-digit basis points to SG&A. Mind you, that every year, the denominator of that calculation, sales, keeps going up. It's rising. I think historically to date, it's probably in the mid to high teens, it's gotten a little more outsized this year because some of the big programs have now been installed.
Notably, at the beginning of this first fiscal year, as an example, our major accounting platform, which is the crux of a lot of things that we'll do on it now, that was $150 million that'll then be amortized over Seven years. We'll keep it longer than that, but that's what we'll amortize it over. That's added to that thing. I think you'll still have incremental costs, and the definition of modernization will evolve also. We keep adding new things to it, rewriting the pharmacy system. There's additional things that we'll do. I think it's going to be less painful going forward. This year is a double whammy because you also have some things that impact sales downward, and so that denominator hasn't grown as fast in that regard. No, I think it's still going to be a drag for a few years.
Much less of a drag than it has been.
Got it. Thanks, guys, and good luck.
Why don't we take two more questions? I think this is our longest call ever, and David and Bob and I are here to continue to answer them outside of this conference.
Again, there are no questions in queue.
Well, thank you very much. Have a good afternoon.
This concludes today's conference call. You may now disconnect.