Tracking the booking progress of ryokan (traditional Japanese inns) in Toyooka City, Hyogo — home to Kinosaki Onsen — for the three-day weekend that includes Labor Thanksgiving Day (Sat Nov 21 – Mon Nov 23, 2026, a public holiday) from 45 days before arrival shows that the first night, Saturday 11/21, had already reached an estimated OCC (based on OTA-listed inventory) of 93.7% at T-45, and had built up to 95.7% by October 26, 26 days out. By contrast, at the same point the eve night (Friday 11/20) stood at only 88.8% and the final night (Monday 11/23, holiday) at 85.5%, meaning more than 70% of the remaining listed inventory across the four nights is concentrated in these two nights. The picture — the first night essentially settled, with only the eve and final nights still in play — is unchanged from what we observed on October 13. In this article we overlay the booking curves for these four nights, the actual shape of the two most recent long weekends, and the seasonal range of estimated settled ADR for Toyooka ryokan, and lay out, with dates, what could already be confirmed at T-30 (10/22, now past) and what needs to be decided at the upcoming T-14 (11/7, the day after the snow crab season opens).
Scope: ryokan category in Toyooka City, Hyogo, N=56–58 properties / 956–982 rooms (fixed cohort based on listings on major booking sites; varies by stay date). Estimated settled ADR covers Toyooka × ryokan, N=77–88 properties. Price metrics in this article are estimated settled ADR (the settled price level estimated from OTA and other sales data, tax-excluded equivalent); occupancy is an estimate based on OTA-listed inventory. Definitions are at the end of the article. Data as of: October 27, 2026.
• ADR (average daily rate) = an estimated settled rate (tax-excluded equivalent), calculated by applying segment-specific adjustment factors to the lowest-plan price level each property publishes on OTAs (per room, double occupancy, tax-included). Checked against property-level results disclosed by listed hotel REITs (91 properties, most recent 3 months), the median error is about 7%. These are estimates and differ from each property’s actual settled prices and accounting figures.
• Area-level ADR (prefecture / municipality) is the median of the estimated settled rates of the properties covered (the level of a typical property in the area).
• Where listed prices are mentioned, they are per-room rates for double occupancy (tax-included).
• OCC (occupancy) = rooms sold as a share of total rooms in the area (an estimate based on OTA sales inventory).
• Source: MetroEngines Research (continuously tracking OTA published prices for approx. 168,000 properties in Japan, of which approx. 27,000 have verified operations)
- — The first night, 11/21, is at 95.7% (26 days out, as of 10/26), essentially settled from 93.7% at T-45. Of the roughly 345 rooms’ worth of listed inventory remaining across the four nights, more than 70% (about 249 rooms’ worth) is concentrated in the eve night 11/20 and the final night 11/23.
- — The final night lands 7.5–12.7pt below the middle nights — that was the pattern in the last two long weekends (the 5-day September holiday and the 3-day October weekend). This time too, the final night is at 85.5% against 94.3% for the middle night, an 8.8pt gap already.
- — Estimated settled ADR for November is ¥31,800 (current estimate, N=77), already at crab-season levels. At T-14 (11/7, the day after the matsuba crab season opens), move the eve and final nights with conditions such as multi-night requirements or opening single-night stays — not with price cuts.
Booking Curves for the Four Nights — The First Night Settled at T-45; What Remains Is the Eve and Final Nights
We start with five booking curves — the four nights of the long weekend (eve night 11/20 through final night 11/23) plus the previous Saturday, 11/14, as a reference — from 45 days before arrival to the latest observation. The scope is 56–58 ryokan in Toyooka City with 956–982 rooms, a fixed cohort tracking the same set of properties for each stay date.
Toyooka City, ryokan, N=56–58 properties. Based on observations from 45 days before arrival to the latest. Source: MetroEngines Research; compiled by HotelBank Editorial Team
The first night, Saturday 11/21, was already at 93.7% at T-45 on October 7 — effectively within the sell-out zone. Over the following 19 days it added only +2.0pt, reaching 95.7% as of October 26, 26 days out. The final night, Monday 11/23 (holiday), by contrast, started at 79.2% at T-45 and gained +6.3pt over the same period to 85.5% at 28 days out, while the eve night, Friday 11/20, rose +6.0pt from 82.8% to 88.8% at 25 days out. Their gains are three times that of the first night, but in absolute terms they remain 7–10pt short of it. The middle night, Sunday 11/22, went from 89.9% to 94.3% (+4.4pt), sitting between the first and eve nights.
Taking fixed points on these five curves gives the table below. “Remaining listed inventory (approx.)” is a rough guide calculated by multiplying each stay date’s total rooms by the unsold share of estimated OCC. The eve and final nights together account for about 249 rooms’ worth — more than 70% of the four-night total (about 345 rooms’ worth). In other words, the two middle nights of the long weekend are now a matter of inventory allocation; the revenue upside lies in the two nights on either side.
| Stay date | Scope | T-45 | T-30 (10/22) | Latest observation | T-45 → latest | Remaining listed inventory (approx.) |
|---|---|---|---|---|---|---|
| 11/20 (Fri) eve night | 56 properties / 956 rooms | 82.8% | 87.6% | 88.8% (T-25) | +6.0pt | ~107 rooms |
| 11/21 (Sat) first night | 56 properties / 961 rooms | 93.7% | 95.3% | 95.7% (T-26) | +2.0pt | ~41 rooms |
| 11/22 (Sun) middle night | 57 properties / 968 rooms | 89.9% | 93.9% | 94.3% (T-27) | +4.4pt | ~55 rooms |
| 11/23 (Mon, holiday) final night | 58 properties / 982 rooms | 79.2% | 83.9% | 85.5% (T-28) | +6.3pt | ~142 rooms |
| 11/14 (Sat) previous Saturday, reference | 56 properties / 963 rooms | 89.1% | 93.0% | 94.9% (T-19) | +5.8pt | ~49 rooms |
| 11/7 (Sat) day after season opening, reference | 56 properties / 963 rooms | 81.0% | 85.9% | 90.9% (T-12) | +9.9pt | ~88 rooms |
Estimated OCC (based on OTA-listed inventory). Remaining listed inventory is approximated as total rooms × (100 − estimated OCC). Source: MetroEngines Research; compiled by HotelBank Editorial Team
The two reference Saturdays are telling. The previous Saturday, 11/14, built up +5.8pt from 89.1% at T-45 to 94.9% at T-19, almost the same level as the long weekend’s middle night 11/22 (94.3%). In other words, for Toyooka ryokan in November, Saturday is a night that fills into the mid-90s even without a holiday, and the first night’s 93.7% at T-45 is better read as “because it is a crab-season Saturday” than “because it is a long weekend.” The other, Saturday 11/7, is the day after the matsuba (snow) crab season opens; it rose +9.9pt from 81.0% at T-45 to 90.9% at T-12 — the largest gain of all the dates shown here. A Saturday right after the season opening may look ordinary at T-45, but it fills rapidly in the final month.
The “Soft Final Night” Pattern in the Last Two Long Weekends — October: Sat/Sun in the 98% Range, Final Night 90.5%
Next, we want to confirm whether the final night really does soften, using the daily results of past long weekends. For the October three-day weekend (Sat 10/10 – Mon 10/12, holiday), estimated OCC for 55–56 ryokan / 965–971 rooms in Toyooka City (based on OTA-listed inventory, observed just before each stay date) was 88.6% for the eve night 10/9 (Fri), 98.9% for the first night 10/10 (Sat), 98.0% for the middle night 10/11 (Sun), 90.5% for the final night 10/12 (Mon, holiday), and 93.9% for 10/13 (Tue). While the first and middle nights were pinned in the 98% range, the final night dropped 7.5pt from the middle night — lower even than the Tuesday after the holiday. Our comparison of this October weekend across four lodging categories in Hyogo, Hyogo Oct Weekends: Holiday Lift Only for Ryokan & Resorts (+5.4pt), likewise confirms that the long-weekend uplift is concentrated in ryokan and resorts.
Toyooka City, ryokan, N=54–56 properties; estimated OCC for October 8–13, 2026 (based on OTA-listed inventory, observed just before each stay date). Source: MetroEngines Research; compiled by HotelBank Editorial Team
The September five-day holiday (Sat 9/19 – Wed 9/23, holiday) had the same shape. Across 51–52 properties / 925–937 rooms, the eve night 9/18 (Fri) was 79.6%, then 9/19 (Sat) 99.0%, 9/20 (Sun) 99.8%, 9/21 (Mon, holiday) 99.9%, and 9/22 (Tue, holiday) 99.8% — four consecutive nights essentially sold out — while the final night 9/23 (Wed, holiday) came in at 87.1%, more than 12pt below the middle nights. Regardless of the length of the holiday, the demand-side behavior of “not staying the final night because the next day is a workday” is consistently observed among Toyooka ryokan.
| Long weekend | Scope | Eve night (Fri) | First night | Middle night(s) | Final night | Middle → final |
|---|---|---|---|---|---|---|
| 9/19 (Sat) – 9/23 (Wed, holiday), 5 days | 51–52 properties / 925–937 rooms | 79.6% | 99.0% | 99.8–99.9% | 87.1% | −12.7pt |
| 10/10 (Sat) – 10/12 (Mon, holiday), 3 days | 55–56 properties / 965–971 rooms | 88.6% | 98.9% | 98.0% | 90.5% | −7.5pt |
| 11/21 (Sat) – 11/23 (Mon, holiday), 3 days *in progress | 56–58 properties / 956–982 rooms | 88.8% (T-25) | 95.7% (T-26) | 94.3% (T-27) | 85.5% (T-28) | −8.8pt |
Estimated OCC (based on OTA-listed inventory). September and October are observations just before each stay date; November is progress as of October 27, 2026. Source: MetroEngines Research; compiled by HotelBank Editorial Team
There is, however, one difference in premise between the past two long weekends and this one. For the September and October holidays, the first night reached 98–99% just before arrival as an actual result, whereas 11/21 is only 95.7% progress at 26 days out. Even so, with the gap to the middle night already at 8.8pt, a scenario in which the final night 11/23 catches up with the middle night at the last minute is hard to envisage based on the past two patterns. Treating the final night not as “a night to wait for it to fill” but as “a night to design how it fills” is the reading consistent with the long-weekend pattern of Toyooka ryokan. Our reverse-engineered analysis of the same Labor Thanksgiving weekend across three lodging categories in Kyoto, Kyoto Nov 21-23 Long Weekend: Business Hotels 98.3%, 75.1% Sold Out, also shows the step down between the first and last days in the same direction.
As a supplement, the day-of-week averages for October 1–27 (53–58 properties / 939–990 rooms) show Saturday at 96.1% and Wednesday at 96.8% as the highest, with Monday at 90.5%, Friday at 91.4% and Thursday at 91.8% on the low side. In Toyooka in October, ahead of crab season, mid-week Wednesday runs as high as the weekend; the fact that there is more than one day-of-week peak is worth keeping in mind as background when thinking about weekday inventory after the holiday.
Estimated Settled ADR for Toyooka Ryokan — December–March Is About 1.6x Summer; November Currently Estimated at ¥31,800
Let us also check the price axis. Overlaying the monthly estimated settled ADR (tax-excluded equivalent) for Toyooka ryokan (category “ryokan,” N=77–88 properties) for 2025 and 2026 clearly reveals the shape of crab season. On a finalized basis, December 2025 was ¥35,900 (N=82), January 2026 ¥34,700 (N=83), February ¥32,900 (N=86) and March ¥32,600 (N=87) — holding in the ¥32,000–36,000 range for four months — while June 2026 was ¥19,500 (N=77), July ¥19,300 (N=79), August ¥22,400 (N=78) and September ¥21,400 (N=79), staying around ¥20,000. The average of the four crab-season months is about 1.6 times that of the four summer months; Toyooka ryokan earn much of their annual revenue in this period.
Toyooka City, ryokan, N=67–88 properties. Solid lines are finalized values; dashed lines are estimates based on current sales (subject to change). Source: MetroEngines Research; compiled by HotelBank Editorial Team
Year over year (finalized months only), January 2026 was +1.2% (¥34,689 vs ¥34,264), February −0.8% (¥32,936 vs ¥33,190) and March −0.6% (¥32,648 vs ¥32,845) — last season’s crab period ran at almost the same level as the year before. In summer, figures swung month to month, with August at −4.2% (¥22,449 vs ¥23,435) and September at +4.3% (¥21,403 vs ¥20,518). For reference, the finalized September 2026 value for ryokan across Hyogo Prefecture as a whole (N=278) was ¥19,100; Toyooka exceeds this by just over ¥2,000.
| Month | 2025 (finalized) | 2026 (finalized) | YoY | N (2026) |
|---|---|---|---|---|
| January | ¥34,264 | ¥34,689 | +1.2% | 83 |
| February | ¥33,190 | ¥32,936 | −0.8% | 86 |
| March | ¥32,845 | ¥32,648 | −0.6% | 87 |
| August | ¥23,435 | ¥22,449 | −4.2% | 78 |
| September | ¥20,518 | ¥21,403 | +4.3% | 79 |
| November | — | ¥31,839 *current estimate | — | 77 |
| December | ¥35,914 | ¥35,206 *current estimate | — | 79 |
Estimated settled ADR (tax-excluded equivalent). YoY is calculated between finalized months only. The 2026 values for November and December are estimates based on current sales; a straightforward comparison with finalized values must wait until month-end. Source: MetroEngines Research; compiled by HotelBank Editorial Team
The estimated settled ADR for November 2026, the month of the long weekend, is ¥31,800 (N=77) based on current sales. It is not a finalized value and may change, but it is already in a range more than ¥10,000 above September’s finalized ¥21,400, stepping up toward December’s current estimate of ¥35,200. This means November pricing is already set at “crab-season levels,” and the first night reaching 93.7% at T-45 happened within this price band. Therefore, when handling the inventory left on the eve and final nights, the question is not “should we cut prices to fill them” but “what conditions should we attach to sell them within the same crab-season price band as the first and middle nights.”
A Reverse Calendar Anchored on the 11/6 Matsuba Crab Season Opening — T-30 Has Passed; T-14 (11/7) Is the Practical Milestone
For snow crab (matsuba crab) fishing in the Sea of Japan, including off Tajima in Hyogo, a Hyogo Prefecture press release (October 31, 2025, for the Reiwa 7 / 2025 season) indicated an opening date of November 6, and the fishing-season table published by Tottori Prefecture also gives November 6 to March 20 of the following year for males and November 6 to December 31 for females (seko-gani). November 6, 2026 falls on a Friday, and as noted above, Saturday 11/7, the day after the opening, had progressed to 90.9% at 12 days out. As we have not confirmed an official announcement for the Reiwa 8 (2026) season at the time of writing, this article assumes the customary November 6 opening. Our article Crab Season 2026: Ryokan Rates Up to 44% Above October Saturdays, which compares Saturday 11/7 across Hyogo, Tottori, Fukui and Ishikawa and examines the rate premium over October Saturdays and the room left to fill, covers the same weekend from the pricing side.
The gap between the season opening and the long weekend is 15 days. What matters in this year’s calendar is that the usual “set prices at T-30” milestone already passed on October 22 (Thu), and that T-14 — November 7 (Sat) — coincides exactly with the day after the season opening. The T-30 values (first night 95.3%, middle night 93.9%, eve night 87.6%, final night 83.9%) showed that over the 15 days from T-45 the first night barely moved (+1.6pt) while the final and eve nights rose +4.7–4.8pt. What should have been checked at that point was which room types your own property’s remaining first-night inventory is skewed toward, and how the sales conditions for the eve and final nights (two-night minimums, whether single-night stays are allowed) are affecting the remaining first-night inventory. If this has not been checked yet, there is real value in getting it done before T-14.
| Milestone | Date | Market observation (Toyooka City, ryokan) | What to check / decide |
|---|---|---|---|
| T-45 (passed) | 10/7 (Wed) | First night 93.7%, middle 89.9%, eve 82.8%, final 79.2% | Confirm the first night settled early. The 14.5pt gap between first and final nights is the starting point |
| T-30 (passed) | 10/22 (Thu) | First night 95.3%, middle 93.9%, eve 87.6%, final 83.9% | Inventory the room types left on the first night and the sales conditions for the eve and final nights. If not done, complete by 11/6 |
| Season opening | 11/6 (Fri) | Day-after Saturday 11/7 at 90.9% at T-12 | News of the first auctions and landings shifts demand sentiment. Log inquiries and last-minute bookings |
| T-14 | 11/7 (Sat) | Latest market values: eve 88.8%, final 85.5% (as of 10/26) | Decide sales conditions for the final night 11/23 (relax multi-night requirements / open single-night stays) and how to handle the eve night 11/20. Make the final allocation of remaining first-night inventory |
| T-7 | 11/14 (Sat) | The previous Saturday itself at 94.9% at T-19 | If your eve and final nights are below the market lines of 88.8% and 85.5%, readjust conditions. For the first and middle nights, prepare to absorb cancellation returns |
| Final days | 11/19 (Thu) – 11/20 (Fri) | In the past two long weekends, the final night landed 7.5–12.7pt below the middle nights | Arrange remaining final-night inventory for last-minute demand. Hold the sell-out zone on the first and middle nights; do not chase occupancy with price cuts |
Estimated OCC (based on OTA-listed inventory), Toyooka City, ryokan, N=56–58 properties. Source: MetroEngines Research; compiled by HotelBank Editorial Team
For Revenue Managers Running Ryokan in Toyooka and Kinosaki Onsen — Implications and Action Plan
Operational insights
1. On the first and middle nights, the question is not “whether to fill” but “what to hold back.” With the market as a whole already at 95.7% for the first night and 94.3% for the middle night, if your own property has reached these lines on the same two nights, there is room to treat the remaining inventory not as a target for discounting but as a buffer for cancellation returns and high-rate last-minute demand. Conversely, if only your property is clearly below the market, it makes more sense to look for the cause in visibility or how room types are presented rather than in price.
2. The final night is at 85.5% across the market too — it is not just your property’s problem. As the actual results of 90.5% for the final night of the October weekend and 87.1% for the final night of the September five-day holiday show, a soft final night is a pattern common to Toyooka ryokan. Rather than breaking price to fill the final night to first-night levels, moving inventory with “conditions” — such as multi-night stays starting from the eve night 11/20 or single-night stays on the final night — makes it easier to protect the crab-season price band (November current estimate: ¥31,800).
3. The eve night, Friday 11/20, has the most upside. Even after gaining +6.0pt since T-45, it stands at 88.8%, still 7pt short of the first night. In a market where the previous Saturday 11/14 fills to 94.9% at T-19, a Friday eve-night stay may keep moving to the very end as an outlet for “demand that couldn’t get into Saturday,” and a design that presents the remaining first-night inventory together with the eve night is worth considering.
4. The 11/6 season opening shifts demand sentiment. That Saturday 11/7, the day after the opening, posted the largest gain (+9.9pt, from 81.0% at T-45 to 90.9% at T-12) shows the power of opening-day news to drive last-minute bookings. With T-14 coinciding with the day after the opening this year, you can sequence things so that you finalize final-night conditions after watching inquiries and bookings for a few days after the opening.
Action plan (with timeline)
| Timing | Action | Decision trigger (tied to figures in this article) | Aim |
|---|---|---|---|
| Today – 11/6 (before season opening) | Compare your own first- and middle-night progress against the market’s 95.7% and 94.3%, and inventory the room types left | If your first night remains below the market’s 95.7%, review visibility and room-type presentation before price | Finish, before the opening, the checks that should have been done at T-30 |
| Today – 11/6 (before season opening) | List the sales conditions (multi-night requirements, single-night availability, meal content) for the eve night 11/20 and final night 11/23 | If the eve night is below the market’s 88.8% or the final night below 85.5%, check whether something in the conditions makes them hard to fill | Line up the “condition” options to move at T-14 in advance |
| T-14 (11/7) – 11/9 | Decide whether to relax the multi-night requirement for the final night 11/23 or open single-night stays | If your final-night progress hasn’t crossed the market’s 85.5% line within a few days of the opening, prioritize relaxing conditions | Move the final night while staying in the crab-season price band |
| T-14 (11/7) – 11/9 | Present the eve night 11/20 together with the remaining first-night inventory (a path to two-night stays starting from the eve) | If your remaining first-night inventory has narrowed to market levels (roughly the 4% range), sell the last first-night rooms via multi-night stays starting from the eve | Fill the high-upside eve night with first-night demand |
| Around T-7 (11/14) | Decide in advance how to release cancellation-return inventory on the first and middle nights without discounting | In a market where the previous Saturday 11/14 fills to 94.9% at T-19, treat returned inventory on the premise that it can be resold within the market price band | Protect rates on the two sell-out-zone nights |
| Final days (11/19–20) | Arrange remaining final-night inventory for last-minute demand, and prepare proposals to shift guests to the following weeks (Sat 11/28 is at 90.2% at T-33) | Based on the pattern of the final night landing 7.5–12.7pt below the middle nights in the past two long weekends, don’t try to make up the final-night shortfall through rate | Accept the final night’s landing as the pattern predicts and link through to the next weekend |
Source: MetroEngines Research; compiled by HotelBank Editorial Team
Summary — Three Yardsticks for Reading a Long Weekend as Four Nights
For Toyooka ryokan, the Labor Thanksgiving long weekend is essentially settled on the first night — 93.7% at T-45 and 95.7% at T-26 — and the remaining issue is how to handle the eve night 11/20 (88.8%) and the final night 11/23 (85.5%). We offer three reusable yardsticks. First, read the progress of a long weekend in two groups: “first and middle nights” and “eve and final nights.” In Toyooka, there was a gap of more than 14pt between the first and final nights at T-45, and it does not close by arrival. Second, in the past two long weekends the final night landed 7.5–12.7pt below the middle nights; build this in as a pattern rather than treating it as an anomaly. Third, with November’s estimated settled ADR currently estimated at ¥31,800 — a crab-season level — move the two shoulder nights with conditions rather than price, and protect the rates of the two sell-out-zone nights. T-30 has passed, but T-14 (11/7), the day after the matsuba crab season opens, is still ahead.
About the Data
- Estimated OCC (based on OTA-listed inventory): OTA-listed-inventory occupancy = 100 − 100 × OTA-listed rooms remaining ÷ total rooms. The scope is a fixed cohort of properties that list at least 30% of their total rooms on OTAs (the same set of properties for each stay date; the property count is the size of that set). This is an estimate based on how OTA-sold inventory is being absorbed and differs in definition from actual room occupancy (rooms not listed are counted as sold, so it tends to run high). Months covered: September and October 2026 (observed just before each stay date) and November 2026 (progress as of October 27, 2026).
- Booking curves: based on observations from 45 days before arrival to the latest.
- Estimated settled ADR: the settled price level (tax-excluded equivalent) estimated from OTA and other sales data (lowest-plan level × segment-specific factor, ensemble across multiple channels). Past months are finalized values; current and future months are estimates based on current sales. Median error against published operating results: 6.6%.
- Scope: estimated OCC and booking curves cover the ryokan category in Toyooka City, Hyogo, N=56–58 properties / 956–982 rooms (each November stay date); the September dates cited (9/18–9/24) cover 51–52 properties / 925–937 rooms, and October covers 53–58 properties / 939–990 rooms (all based on listings on major booking sites). Estimated settled ADR covers Toyooka × ryokan category, N=77–88 properties (varies by month; estimates for January–March 2027 cover 67–76 properties), and ryokan across Hyogo Prefecture N=278 properties (September 2026).
- Days of the week are based on the 2026 calendar. Monday, November 23 is Labor Thanksgiving Day.
- Data as of: October 27, 2026. Sales and inventory change daily, so the figures in this article are a snapshot at the time of retrieval.
References and Sources
- Hyogo Prefecture, “Status of Snow Crab Resources for the Reiwa 7 Fishing Season” (press release, October 31, 2025; Japanese): https://web.pref.hyogo.lg.jp/ns01/20251031.html
- Tottori Prefecture, “Resource Management of Snow Crab” (males Nov 6 – Mar 20, females Nov 6 – Dec 31; Japanese): https://www.pref.tottori.lg.jp/305461.htm
Related Articles
- Crab Season 2026: Ryokan Rates Up to 44% Above October Saturdays
- Kyoto Nov 21-23 Long Weekend: Business Hotels 98.3%, 75.1% Sold Out
- Hyogo Booking Curves: 3.6x Pickup Gap, Ryokan Sellout Rate 28.4%
- Yufuin Ryokan Oct Long Weekend: Final Day Trails Day 1 by 16.3pt
- Hyogo Oct Weekends: Holiday Lift Only for Ryokan & Resorts (+5.4pt)
- Crab Season 2026 Opens Nov 6: 613 Inns, Rates Up to 3.6x in Echizen
- Hyogo’s Hotel Market in 5 Layers 2026: ADR from ¥10,700 to ¥29,600
- Nov 21-23 Foliage Weekend: Kyoto Ryokan 18% Rooms Left, Nagano 32%