open All Channels
seplocked Science and Industry
blankseplocked How to value goods for a manufacturer-station trader broker agreement
 
This thread is older than 90 days and has been locked due to inactivity.


 
Author Topic

Kethas Protagonist
Protagonist Ventures
Posted - 2011.05.29 06:18:00 - [1]
 

I run a hisec industrial (primarily manufacturing) corp that, for now, is based outside of Caldari space. We've expanded to the point where, because of the higher volumes, we need to both buy a decent chunk of our materials and sell a decent chunk of our product in Jita, rather than simply participating in the local trade hub.

Currently, after making the trip to Jita, I'm just putting up buy and sell orders myself, babysitting the former until they fill, and then leaving. This is undesirable, though, for a bunch of reasons:

-Every minute I spend in Jita babysitting orders I'm not spending on other activities more central to our industrial corp, or babysitting orders back at our normal trade hub.
-My Caldari faction standing isn't as high as the others, so I'm paying higher broker fees.
-I have to buy all the materials I want for the day before leaving Jita. This means I tend to temporarily drive up prices and pay a premium for my sudden, large buy orders.
-I can't monitor my sell orders while I'm outside The Forge. This means I often have sell orders sitting undercut and inactive for hours or days at a time.

Our corp knows a few people that earn their income not through manufacturing but through station trading in Jita. Since they're trading in Jita already, we'd like to hire them as brokers: we arrive in Jita with finished goods, trade them to the broker(s) for pre-purchased materials and some isk, and can then immediately head out. We'd pay them a premium on Jita buy for materials and offer a discount under Jita sell for finished goods, the difference being their profit. It's mutually beneficial.

I've reached an impasse with the first such potential broker I approached. We're both convinced it's a good idea and would like to go through with it, we just can't figure out how to set a "fair" price for the materials and finished products when we trade. (I'll walk through the thinking behind valuing finished goods, but valuing raw materials poses similar questions.)

-----

The basic idea: we want to design a pricing scheme for finished products that offers a profit both to my own corp (the manufacturer) and the broker. A reasonable floor on that price is:

<valuation of raw materials that went into the product, at Jita buy>
+ <the broker fees I would incur on such buy orders>
+ <NPC station mfg slot fees, POS fuel costs, etc.>

That's the price that would simply cover my costs (valuing my time as free, incidentally) and give all the profit to the broker. A reasonable ceiling is:

<valuation of finished goods at Jita sell>
- <the broker fees and sales tax he'd incur on such sell orders>

At that price, the broker would just barely be able to recoup his costs, assuming he was able to sell all his stock at the current Jita sell price; all the profit would go to me, the manufacturer.

-----

If that were it, agreeing on a price would be pretty easy; we'd negotiate a split of the profit (say, 2:1 in favor of the manufacturer + hauler) and value the finished goods accordingly. It isn't that simple, though. The current Jita sell price isn't an unbiased estimate of the price that the broker will be able to get; we're dumping stock on the market, in some cases a lot of stock, so we can expect Jita sell to go down. We thus both agree that the pricing scheme above is "unfair" to the broker; it overestimates his profit. We don't know how to correct the pricing scheme to adjust, though - it seems to require estimating how far the price will drop when our stock hits the market, and we don't know how to do that, certainly not in an automated/spreadsheet fashion.

-----

Does anyone have any advice on how we can agree on a way to value the finished goods (and raw materials) we're trading? The best approach I've thought of would be using averaged-out historical data instead of just the current price, but a) that implicitly makes the broker hold onto stock for longer, and b) you're trading price "unfairness" for exposure to price fluctuation.

Kethas Protagonist
Protagonist Ventures
Posted - 2011.05.29 06:23:00 - [2]
 

A couple side notes:

1) Sorry for the wall of text.

2) The cleverer among you might have noticed that I'm valuing materials and products in a circular fashion, i.e. I'm valuing products in terms of the value of materials, and materials in terms of the value of products. I haven't thought of a solution.

3) If anyone would potentially be interested in employment as an above-described broker at a hub other than Jita (or Jita too, I suppose), please hit me up with an ingame mail.

Shira Elan
Posted - 2011.05.29 06:46:00 - [3]
 

Join the bulk trade mailing list and try selling your stuff there (assuming you are making a lot of it). For reference, T2 modules often sell for a 30-50k discount per unit to Jita prices. The average of Jita bid and ask also often does the trick.

That seems much easier than finding a single broker who won't always be online, etc.

Cyniac
Gallente
Twilight Star Rangers
Posted - 2011.05.30 23:51:00 - [4]
 

Good luck with this I tried to pursue this approach as a manufacturer with several Jita station traders coming up with various valuation schemes but ultimately could not find a trading partner / profit scheme which could work long term.

Originally by: Shira Elan
Join the bulk trade mailing list and try selling your stuff there


And this here is the bulk of the problem - Selling stuff is easy but when you also need to spend time babysitting buy orders (and lots of buy orders at times) why would you deliberately lose out on the profit if you *still* are bound to a station to buy your goods? Manufacturers want to either be freed from the trade hub syndrome or will simply have to deal with it - half a deal is not a good deal.

As for details of the valuation I can just wish you the best of luck - to a large extent trust between you and the broker would work wonders but ... not as simple as it sounds.

Kethas Protagonist
Protagonist Ventures
Posted - 2011.05.31 13:21:00 - [5]
 

Originally by: Cyniac
...Why would you deliberately lose out on the profit if you *still* are bound to a station to buy your goods?


That's one good reason why simply mailing Bulk Trade each time I have a cargo I want to unload isn't a good solution. (Sending Bulk Trade a mail saying "Hi, I make X Y Z and need materials A B C, I'm looking for a regular broker, please contact me if interested" could very well be a good way to find people, possibly with more experience in broker agreements like this than I have, so that's worth doing. It's on my list of things to do, but unfortunately right now that's a rather long list.)

The other reason is that using the mailing list to find buyers wouldn't address the core issue in the OP - figuring out an accurate, automated way to fairly value sets of goods. Let's say I mail the list and say "I have 10 X, 50 Y, and 100 Z for sale." That's nice, but it's irrelevant to figuring out how to price them. I have to still look at the market and guess what kind of price I can get for them each time I have a cargo to sell. If I guess wrong - and I *will* guess wrong sometimes - I'll need to change my prices. I'm back where I started: instead of setting up and babysitting a sell order on the market, I'm setting up and babysitting a sell order on Bulk Trade.

-----

Fortunately, I think I've come up with a possible solution (or at least an approach that's a significant improvement over just using Jita sell as in the OP). First, let me remind you of the overall goal here: we're trying to take an arbitrary cargo of goods and quantities for sale, use the information at our disposal (the state of the market and the nature of the cargo) to get an unbiased estimate of the price the broker will be able to get, and then set a value on the cargo that splits the profit according to some negotiated ratio. That's it.

To understand why I can't simply use Jita buy as an estimate for how much my cargo is worth if I want the cash "right now," you also need to know that what I make almost never trades on buy orders. >99% of the trades (both in number and isk value) in Jita are consumers purchasing small amounts off sell orders, and the buy orders are largely fishing orders (i.e. they're priced below mfg cost). As a result, few people sell to buy orders, a claim that's publicly verifiable by noting the relevant Donchian channels are a) narrow and b) include sell order prices but usually not buy order prices. If Jita buy orders were competitive, of course, I could just sell to them and skip this entire hunting-for-a-broker bit.

-----

If we just use Jita buy as our estimate for the value of the materials that went into the products and Jita sell as our estimate for the value of the finished goods, like in the OP, the formula's pretty simple. Let's say that the broker and I agree to split any profit 2:1 in my favor.

<Price the broker pays me for the full cargo> = <Mfg cost (1)> + (2/3) * ( <Cargo value at Jita sell (2)> - <Mfg cost> )

(1) Mfg cost: value of materials at Jita buy + broker fees on such buy orders + mfg slot fees + POS fuel + etc. etc.; may or may not also include an estimate of the value of my time spent hauling, expected time-averaged cost of loss of cargo to suicide gankers, etc.

(2) Cargo value at Jita sell: revenue from sell orders at current Jita sell, less sales tax and broker fees on such sell orders

to be continued, much like America's budget deficit

Kethas Protagonist
Protagonist Ventures
Posted - 2011.05.31 13:49:00 - [6]
 

So far, we're using two bits of public information to estimate the value of the cargo (Jita buy for materials and Jita sell for finished goods). That's nice, but as I point out in the OP it doesn't paint the whole picture.

If widgets cost me 100 isk to make and sell in Jita for 400 isk, then according to the simple formula above I could sell as many widgets as I wanted to the broker at a constant 300 isk. What if only ten widgets trade a day in Jita, though? It's probably reasonable for me to sell him one widget at close to 300 isk, but it's not fair to the broker for me to unload a thousand widgets on him and expect 300k. He's going to have to lower his price in order to get the widgets to sell.

We need to incorporate another piece of public information: the proportion of Jita's daily volume, for each particular good, that I'm trying to sell the broker in a given cargo. If that proportion is low, then the broker probably won't have to lower his price far below Jita sell, so our valuation should stay close to the simple formula above. If that proportion is high, though - if I'm trying to sell him a significant fraction of Jita's daily volume - he's going to have to lower his price more to make them sell, though, so it's only fair for me to ask a lower price from him.

My approach - which is admittedly simplistic, but I haven't thought of a better one - is to fit the broker's expected profit per unit to an exponential decay curve. Take a look at this plot. It has "percentage of Jita daily volume I'm trying to sell the broker" on the x axis, and "percentage of profit, at CURRENT Jita sell, that we predict the broker will be able to earn" on the y axis. This particular curve assumes that for every full 100% of Jita's daily volume that I sell the broker, he'll have to cut his profit (not his price, his profit) by 75%, but you can plug in different numbers and get different curves.

So for each type of product in the cargo, the formula becomes:

<Price the broker pays me for the full cargo> = <Mfg cost (1)> + (2/3) * <Volume penalty (3)> * ( <Cargo value at Jita sell (2)> - <Mfg cost> )

(1) Mfg cost: value of materials at Jita buy + broker fees on such buy orders + mfg slot fees + POS fuel + etc. etc.; may or may not also include an estimate of the value of my time spent hauling, expected time-averaged cost of loss of cargo to suicide gankers, etc.

(2) Cargo value at Jita sell: revenue from sell orders at current Jita sell, less sales tax and broker fees on such sell orders

(3) Volume penalty: 0.25 ^ <Proportion of Jita daily volume in the cargo>

This means that if I, the manufacturer, screw up/guess wrong/otherwise end up with a lot of stock and show up in Jita wanting to sell a huge number of items relative to the daily volume, I offer the broker an equally huge discount, in an easy to understand and automated fashion.

-----

Let's walk through a couple examples so you can get a feel for how this works.

Widgets cost me 1M isk to make (including everything - materials, broker fees, etc), and currently sell in Jita for 1.5M (after sales tax and broker fees). 100 widgets move a day. The broker and I agree to split profit 2:1, and we also agree to estimate that for every full 100% of Jita volume I dump on him, he'll probably have to cut his profit by 75%.

First, if I only sell the broker a small number of widgets, the volume penalty should be small. Let's say I sell him five widgets.

<Price I charge the broker per widget> = <Mfg cost> + (2/3) * <Volume penalty> * <Jita sell - Mfg cost>

= 1M + (2/3) * (0.25 ^ 5/100) * (1.5M - 1M)

= 1.311M per widget. My profit is 311k per widget; we estimate that the broker can sell them at 1.467M per, giving him a profit of 156k per widget.

still to be continued, because you can't stop the Cole Train

Kethas Protagonist
Protagonist Ventures
Posted - 2011.05.31 14:07:00 - [7]
 

Next, if I show up in Jita with a huge cargo of widgets, the volume penalty should be larger, and I should be forced to offer the broker a discount. Let's say I sell him 200 widgets - two days' worth of Jita volume.

<Price I charge the broker per widget> = <Mfg cost> + (2/3) * <Volume penalty> * <Jita sell - Mfg cost>

= 1M + (2/3) * (0.25 ^ 200/100) * (1.5M - 1M)

= 1.021M per widget. My profit is 21k per widget; we estimate that the broker has to tank the market and sell them at 1.031M per, giving him a profit of 10k per widget.

-----

I've thus improved my potential station-trading broker agreement to offer a discount if/when I sell so much product that the broker will have to significantly lower his price to get them to sell. My goal is to design a contractual agreement that both I and the broker(s) believe is fair, and that lets me quickly and easily value whatever cargo I happen to show up in Jita with.

Can anyone think of additional factors that a) are publicly known and thus can be incorporated into our formula, and b) affect the price the broker will be able to get? One thing I'd like to improve is that while we now penalize me, the manufacturer, for selling large quantities of the same product, we don't similarly reward me for consuming large quantities of the same raw material. If my manufacturing jobs require 50% of Jita's daily volume in widget parts, then the Jita buy for widget parts when I show up in station and am preparing to buy my daily amount off the market isn't accurate - it estimates my costs as incorrectly low. I (or my broker, acting in my stead) will have to raise my bids above current Jita buy in order to fill my buy orders.

Durin Sarga
Posted - 2011.05.31 18:22:00 - [8]
 

Why not do a markup + profit share scheme?

Widget costs X in manufacturing cost.

You sell to broker at 1.1X
Broker sells on market for 1.2X

IF the widget sells on market for 3.0X then you split the profit 50/50.

Everyone is guaranteed a 10% markup and a share of the additional profits. If the broker is unable to sell the widget at 1.2X then he eats the cost of the stock he purchased from you at 1.1X and then informs you of the price change.

Require the broker get audited or have his accounts be reviewable by you 'on-call' to make sure there is accountability in the system. Obviously the markups in the above example are just for demonstration purposes and should be negotiated between the broker and yourself.

RAW23
Posted - 2011.06.01 17:56:00 - [9]
 

I think the suggested 2:1 split in favour of the manufacturer/hauler is weighted rather too heavily in your direction. In my experience the absolutely critical thing in such arrangements is to keep the station trader sweet, as by cutting out the management of the sell orders you are probably reducing your keyboard time by 50% or more. The last thing you want is to negotiate a deal that is favourable to you and then find that the trader pulls out after a week or so. My suggestion, and what I have done in the past, is a simple 50/50 split on the profits.

On a related note, if you can find a trader you trust the preferred solution would be to bring him into the corp as you can then squeeze some extra value out of the arrangement by getting him to manage your buy orders for you as well (you may have covered this; I skipped through much of the wall of text Wink). You can then use the delivery hangar as a pseudo-corp hangar as well, which means you won't have to worry about synchronising your play times too much to organise trades with each other. You will also save a great deal of administrative time as you won't need to work out production costs and sales values for each batch, just add up your NAV at the end of each period (day/week/month/whatever) and divide everything above the previous week's NAV between you. Generally, an integrated sales/manufacturing operation is much easier to scale, which sounds like it could be an important point for your corp as time goes on. If you have someone in mind who you trust a bit but not entirely then you can restrict access to corp assets and ask them to lodge a surety bond with either you or a trusted third party.


 

This thread is older than 90 days and has been locked due to inactivity.


 


The new forums are live

Please adjust your bookmarks to https://forums.eveonline.com

These forums are archived and read-only