How Much Do Manila Prostitutes Earn

The good news is my wife is from the Philippines. The bad news is my wife is from the Philippines.

Because, in addition to my wife, I LOVE prostitutes. Just read my other stories on this site--they're all about hookers one way or another. Despite (in my opinion) that the best place for prostitution in the whole wide world is the Philippines, because of my wife I am prevented from partaking.

Travel to the Philippines for me is very frustrating. The best I can do is window-shop. Fortunately, through the miracle of the Internet I can do that in style. Indeed, it's actually turned into a research project--full of facts, figures and data, some of which might actually be meaningful. All of it is fun.

This essay is a report on that research project.

This research is based on data taken from a well-known website. Literotica rules prohibit me from providing a link or from advertising them. I interpret this as meaning I can't quote their name here. However, it is based in Bangkok and has a name that is roughly a synonym for "kiss." It will hereafter be referred to as The Website.

The question is: How much do prostitutes in Manila earn?

Alternative ways of answering the question:

1) Ask the whores. I have no way of doing that. If I did, I have no reason to expect an answer. And if I got an answer I have no reason to believe it's true. Beyond which I expect many hookers don't keep an accurate tally of their receipts. They probably don't even know themselves.

2) Ask The Website from which I draw information. Again, I have no credible standing to do that. I have no reason to expect they'd honor my request even if I made it.

3) Let The Website do the research. This is obviously an improvement over what I can do. They have direct access to all of their data, unlike me. But at the end they would still need to make assumptions about how women spend their time. Perhaps they could do it better than I can, but maybe not that much better. Either way, The Website ain't doin' the project.

4) Use a more sophisticated computer program to scavenge data directly off the web, rather than copying it down by hand. That really would be an improvement. Indeed, one could then get almost as much info as The Website has in their database. But I don't have the skill set for that, and the learning curve is too steep.

So you're stuck with me, warts and all. That said, I think I have compiled as good an estimate of what a Manila prostitute earns as exists anywhere. Plus I can tell stories.

The Website

This research is based on data drawn from a website. The Website is a relatively new company that can be described as Uber for Whores. It was founded in Bangkok and has since taken off in Manila. They are attempting to enter other markets, but so far with limited success. Who cares--here we're only interested in Manila.

The company is app-based. Hookers post ads and customers can respond to them. If the customer pays a fee he can chat with the ladies (assuming she is willing), but otherwise all he can do is "book" them. To do so requires some personal information:

--his name as it appears on his hotel reservation;

--email address;

--name of hotel or condo;

--room number;

--special requests;

--willingness to pay a special fee, e.g., for A-level.

Presumably his phone number is forwarded automatically through the app.

That's all hard to fake--a prankster in Pakistan is gonna have a hard time simulating all that data. But it gets harder because The Website gets your location off your phone--they know if you're telling the truth about your hotel. Which makes it nigh impossible to lie.

That means when a girl gets a The Website booking it is nearly certain to be serious. She's not inundated with jokers and thrill-seekers. The results are benefits to you. First, most girls promise to respond to the booking request within 10 minutes. No more waiting around for an hour or more till she gets back to you. And second, while still protecting her privacy, her phone transmits her location and estimated time of arrival to yours. You know where she is.

So while I can't make a booking from my home halfway around the world (the app tells me I'm currently 13,587.6 km away from Paula even as I write this), I can see in real time which prostitutes are for sale in Manila right now.

Presumably, as soon as she accepts your booking, her ad is taken off-line. She'll receive no more booking requests for as long as she's attending to you. When you're done with her she puts her ad back on-line, along with the promise to respond within 10 minutes.

When a girl is on-line, i.e., her picture shows up on the website, I count her as "available." If she goes off-line, I assume she's "booked," i.e., attending to a client. By looking at the on-line/off-line pattern I can estimate how many customers she has, how long each of her dates are, and roughly how much money she earns. Obviously making all those estimates requires some assumptions. I'll detail those below.

Most hookers follow the 10-minute rule. Some claim only to "usually respond within 10 minutes." A few settle for "rarely responds within 10 minutes." My experience is that the "rarely" girls are rarely booked--and for good reason. I, for one, am not gonna hang around waiting for some silly slut to get back to me--especially since there's so much competition just outside the door. But the "rarely" girls mess up my experiment because they may not go off-line when they have a client.

About half the girls are independent, not associated with an agency. Presumably they answer their own phones, and when they say they'll get back within 10 minutes, it's them that actually gets back.

The remainder are affiliated with agencies, so possibly the agency returns the phone call. Likely they contact the tart right away--at least that's what we'll assume. An agency that can't get a commitment from her within the deadline isn't worth very much.

As said, the research project, like any other, involves assumptions. The first one is that all girls are gonna answer within 10 minutes. Now that is probably true for 80%, and for the remainder it's mostly "usually," which is close enough. Only a few admit to "rarely" (and I believe they have to pay extra for the privilege). In retrospect I should have included that information when I initially constructed my spreadsheet, but I didn't. In point of fact I don't think it makes much difference. As said, the "rarely" girls are rarely booked, and they slide off the site soon enough.

(This research was done over 5 weeks during April and May, 2019. I had been following The Website for several months prior to beginning this project, albeit not writing anything down.

The Website, before listing a companion, requires a photo id and verifies her age. (See their Terms & Conditions page, paragraph 3.5.3). On that basis I affirm that all people mentioned in this post are over 18 years old.

The Website has a separate section for ladyboys. I ignored that section. This research is only about natural born females. There is some cheating. More than a few "girls" make only one appearance on the site. Presumably they are kicked off for violating some rule or another, e.g., not being a girl.

Some girls use agencies, the names of which are posted on The Website. Those have webpages, and many of the girls mentioned here are posted on those pages. The prices on the agency pages are typically higher than the The Website price.)

The Website provides a list of women (along with vital statistics, prices, and in most cases a commitment to respond to a booking within 10 minutes) who are available for sexual services in Metro Manila. This list is updated in real time. Along with putting their names into a spreadsheet, I included the price for a two hour date. (The Website lists prices for as long as 24 hours.)

The Website earns money by charging fees for premium services. Fee-paying clients get access to customer reviews (handled the same way as Uber reviews), along with access to the chat feature, and some other bennies. Fee-paying hookers are not limited in the number of bookings, and are not required to follow the 10-minute rule. Unlike an agency, The Website does not take a percentage of the girl's earnings.

While details are given in the next section, a summary of the research protocol is here:

I visited The Website four times per day. Girls listed were marked as "Available." Girls in my database, but not on the website were assumed to be with a client, and were marked as "Booked." However, when a girl disappeared from the website for 48 hours or more, she was assumed to be taking time off--she was marked as being on "Vacation." Even hookers have to sleep, and so I marked some times as "Sleep." Often girls new to my database appeared on the website, and they were marked as "New" for all readings prior to their initial appearance.

Girls marked as "booked" are with a client, and by counting the number of "booked" marks I can estimate how much money she earns. Girls are considered "on duty" if they are booked, available, or sleeping. (Like you, they don't quit their jobs when they sleep.) Girls who are "New" or on "Vacation" are not "on duty." Since I took readings four times per day, the number of "Duty Days" is the on-duty value divided by four.

Money:

US$1 = 50 Philippine pesos = P50, or close enough.

Thus P5000 = $100, P10,000 = $200, etc.

As of this writing $1 = P52.3, though during my research the dollar generally was in the P51, range, which is the value I used in my calculations.

Procedures & Assumptions

(This section can be skipped by readers not interested in the details. It gets technical.)

I checked the website four times a day, at approximately midnight, 6am, noon, and 6pm. I recorded the date and time to the nearest hour. Times were recorded in Eastern Daylight Time, which helpfully is exactly 12 hours behind Manila. So the corresponding times in Manila are noon, 6pm, midnight, and 6am, respectively. In this report all times are given in Manila time, and reported as noon, evening, night and dawn.

The project began on April 11th and concluded on May 15th, 2019.

Each row on the spreadsheet lists the name of a girl and her price. Columns are headed by date and time, e.g., 4/15-18, which means April 15th at 6pm EDT, which in Manila is April 16th at dawn. Since I wasn't always right on schedule when I took a reading, any time between 1500 and 2000 EDT was taken to be dawn. But I was pretty good--by far most readings are within half an hour of 1800, and most of the rest are 1700 or 1900. Only a few are further afield than that.

Similar caveats hold for the readings at noon, evening and night.

Over the course of the month I missed three readings altogether. Minor adjustments in procedures were made to accommodate the missing data.

Women listed on the webpage received a "0", which means they were available. Over the month I accumulated a database of 163 women who have been listed as available on The Website at least once, that is, gotten at least one zero.

Of those 163 ladies, the default assumption was those who were not listed had a booking. They were labeled by entering a "1", meaning "booked." But of course it's not that simple--girls can turn off their app for all sorts of reasons: childcare, sleep, a day job, or because they don't want to be on The Website anymore.

So somehow I have to distinguish between women who are booked (and deserve their "1") and those who've just turned off their app. Of course I can't do that with any precision, but I estimated it using the following rules.

1) A new girl just added to The Website did not receive a 0 or 1 for days prior to my encountering her. Instead I entered an "N" for all readings prior to her arrival. Of course she may have been on The Website before I met her, but I assumed she wasn't. At the very start of the project (4/11-0700), all the girls were listed as available, none of them were booked and none of them were new. In short, at the beginning my database wasn't big enough to make any conclusions. For that reason the first 48 hours of data is excluded from the calculations. The results are reported only on data beginning April 13th, evening. For that reading I had 80 girls in my database, of which 38 were available and 17 were booked. The ratio between available and booked has remained roughly 2:1 ever since, though the absolute number of whores has increased with time.

2) A girl who had not shown up as available on The Website for 48 hours was assumed to have turned off her app. That is, if girl had eight "1"s in a row, I assumed they weren't bookings but simply vacation time. So I changed all those "1"s to "V"s, for Vacation. All future entries are V unless at some point she becomes available again (i.e., gets another 0). Girls can be put on vacation 8 readings past the current one, which effectively reduces the number of booked girls. By the end of the project approximately half the girls are on vacation at any given time.

3) Even sluts have to sleep. The most common time for sleep is dawn thru noon. That one escort agency stops answering the phone after 3am makes my point, and they don't start again until noon the next day. Not taking this into account leads to the counterintuitive result that the busiest time for a prostitute is dawn. That's not true--unless they have an all-night booking, few prostitutes are fucking at 6am. So I assigned an "S" (for Sleeping) to girls who had a "1" at dawn, but with a "0" the previous midnight. Similarly, I assigned an "S" at noon if she got an "S" at dawn, but had a "0" the following evening. Or put another way, ladies who disappeared from the website at only dawn or noon (not including before or after) were assumed to be sleeping. This was typically about 10 women every dawn, and fewer at noon.

4) I kept a running sum of numbers in a girl's row. Obviously zeroes add nothing, and neither do Ns, Vs, or Ss, so really it just counts the number of ones. That total, which we'll call the Sum, is a proxy for how much time she spends with customers. Of course it's not exact, but it's the best I can do for you.

5) I kept a running "On Duty" tally. All readings containing a 0, 1, or S are counted as "on duty." Readings showing N or V are off-duty. While there were technical reasons why I included S in the on duty figure, it also makes sense. Nobody quits their job just to take a nap. The number of "Duty Days" is calculated by taking the "on duty" number and dividing by four (since there were four readings per day). The maximum possible number of duty days is 32.

So every girl is assigned a 0, 1, N, V, or S for each reading. The column headed 5/8 at dawn (reasonably typical) had 25 "1s", 37 "0s", 7 "Ns", 78 "Vs", and 16 "Ss". Those numbers add up to 163, which is the total number of whores in the final database. All columns will add up to the same number--the number of Ns will decline as time goes on.

I took 20538 readings over the course of the experiment. Of those, 2240 were "1s", 6049 were "0s", 8314 were "Vs", 3518 were "Ns", and 417 were "Ss".

The only hard data are the zeroes. That's the number of women actually listed on The Website at a particular time. Everything else is an estimate based on the rules above, i.e., inferred from the girls who do NOT show up on the webpage but exist in my database. But it seems qualitatively accurate, though obviously wrong for some individuals.

One flaw in the design is I did not anticipate that the ladies would change their prices. I had no good way to record that information. After the first week I started a separate spreadsheet that recorded price changes. Most girls never changed their price. Most price changes were within P1000 ($20). A few girls changed their prices often--sometimes several times per day. A more sophisticated analysis would have taken this into account. The price used for calculations here is her price at the end of the experiment.

So now given this pattern of ones and zeroes, how can I estimate her earnings? Again, I have to make a bunch of assumptions.

Strings of ones can range from a single "1" to being seven "1s" in a row. (Any string longer than that turns into Vacation, for which no income is credited.) I assume each sequence corresponds to one customer. Of course this can't be entirely right. Some girls may get two customers back to back (I think this is rare), or more commonly they may be taking some personal time. There is certainly some of the latter--it typically takes them 2 hours to prepare and travel to the client's location.

Here are the rules:

1. Let "price" refer to her listed price for a 2-hour date. (Again, all calculations were based on the price as of the end of the experiment.)

2. If she has a single "1", she receives her price, i.e., a two-hour date.

3. If she gets two "1s" in a row, she receives double her price, e.g., for a 6-8 hour date.

4. If she gets three "1s" in a row she gets 2.5 times her price, e.g., for a 12 hour date.

5. If she gets four "1s" in a row she gets three times her price, e.g., for a 24 hour date.

She receives no credit for the 5th, 6th, or 7th "1" in a string. These are assumed to be personal time off.

The assumption is that for most girls, double her price bought you six to eight hours with the lady. And for triple you got her for the full 24. While girls often advertised their willingness to spend a weekend with you on Borocay Beach or someplace, that likely doesn't happen very often. The Website lists no prices for longer than 24 hours. Typically that price is triple the 2-hour cost, but often it was more or less.. If I had recorded prices for all time-points I could have calculated this more precisely, But I didn't so I can't.

Some girls (especially the higher-priced ones) did not offer themselves for the whole 24, limiting dates to as short as 3 hours. My algorithm does not take this into account.

The attentive reader will immediately suggest that the whores don't always get their asking price. Indeed, The Website (selling access to their chat feature) all but invites clients to negotiate the price. The discount obviously varies, but is likely no more than 15%. But pick whatever number you want and deduct it from my results. I'm not making any more assumptions.

I'll mention a confounding event that clearly impacted my data. My research window included Holy Week (4/14 - 4/21 -- the week ending Easter Sunday), which is a major holiday in the Philippines. People go home to the provinces to spend time with their families. This undoubtedly affected my statistics.

Results & Discussion

As mentioned, there were 163 girls in the database by the end of the experiment. Of this, 29 of them earned zero money, meaning that 134 ladies made at least some money from The Website during the experiment (according to our algorithm and assumptions). Of the 29 non-earners, a large fraction of them appeared on The Website only once or twice. I presume they were kicked off because of some rule violation. Ten of the non-earners had 7 or more duty days, while 15 had less than one. (The maximum possible number of duty days is 32.)

The remainder of this section considers only the 134 women who earned more than zero.

Of the 134 earning prostitutes, the highest earner by far was a lady name Jewel, who according to my algorithm banked $11,177. (While I report prices in pesos, earnings are given in US dollars, which is likely more intelligible to both the reader and me.) She worked more than 28 duty days and serviced 26 customers. Of course this is estimated via my algorithm, but I think it is most likely to be true for Jewel. Her pattern of 1s and 0s matches my model nearly exactly. Jewel's income per duty day comes out to nearly $396.