There are doubtless some game mechanics of value to be built here - I do not have time to tackle them myself but I’d encourage you to explore it further…!
This is the kind of thing I was thinking about as far as limiting them.
In addition to them possibly being unavailable, I can’t help but think that overusing them would lead to serious penalties to domain morale and incite massive police actions.
If your criminal syndicate of 100 men is assassinating 100 people every month (1200 per year) in a class IV market (625-2499 families), your guild alone would confer a murder rate of between .384 and .096 murders per capita. Per Wikipedia, the city in the world with the highest murder rate in 2011 had a murder rate of .002 murders per capita. (The 50th had a rate of .0003).
In other words, if your guild is concentrating purely on assassinating, spying, and other high-mayhem crimes, it is going to qualify as a crime spree of legendary proportions. The city will become known as an unsafe place to be and people will start fleeing it as they believe the crime syndicate to be in control.
Though that said, I think the easiest solution might be to simply enforce the average hijink income table. Anything that you don’t have a hook for generated in actual play (like finding a patron who wants you to spy on someone or assassinate someone), just use the average.
And if the domain rulers are complicit in letting the thieves’ guilds off without punishments when caught, well, that’s just asking for rebellion, outside interference with the support of the local people, or adventurers and heroes coming knocking (including ambitious 9th-levels who see a potential domain to liberate & rule) …
…and that’d be the perfect way for a thief running a guild (or someone who is the henchmaster of that thief) to take over the domain proper.
Institute general panic and then come in as a savior.
I quite like this idea. It would probably force some diversity of hijinks, which is both a good and bad thing, in that it’s realistic but increases administrative overhead / time to resolve compared to just putting everyone on one or two hijinks (as I usually see).
We always assumed that profits from spying were primarily from blackmail rather than an outside funding source, though.
Good point; if you’re blackmailing someone with every secret that you spy, then every time you perform a spying hijink, your guild makes an enemy.
After even thirty or forty such hijinks, given that the people you’re spying on are rich and therefore powerful (both in terms of personal and political power), you might start to be in trouble. After a hundred? After a thousand?
Even if the city can support having enough nobles to blackmail, eventually, you are going to end up with someone (either the domain ruler or the nobles themselves) bringing an army into the streets to wipe you out.
Finding a patron to spy on their enemies strikes me as safer; the patron draws the heat, and the target doesn’t necessarily know that they were even spied upon by thieves.
I suppose this is the point where I should bring up that I’m an economist (or, at least, that’s what my degree claims)…
Just a curiosity, but what is the source of the income inequality distribution you used for the empire?
ACKS is built on a very large spreadsheet that takes into account the total wealth produced by the society and the amount of wealth required to reach various levels and rule various domains, as well as the average distribution of the lordships…
Yeah, what you’ve managed to produce is incredibly impressive. I was just very interested in the wealth distribution you quoted:
The bottom 77% own 15% of the wealth.
The top 23% own 85% of the wealth.
The top 5% own 75% of the wealth.
The top 2% own 55% of the wealth.
I was wondering whether this distribution was primarily a driven by the rules for level progression and domain rule or whether it was a snapshot of a particular century/region. Or even the amalgamation of several sets of inequality data, i.e. classical, feudal or modern. For the modern United States I know that the amount of wealth owned by the top 1% has been hovering between 33-38% the last 20 years. For 2nd Century Rome I’ve seen estimates as low as 1% only holding 16% of all wealth, though I always assumed it was actually more like what you posted. I’m interested to know how easily historical projections can be imported into your setting or whether ultimately its more a combination of historical data for your bedrock numbers followed by the logic of game progression.
Ah! It is the latter - I used historical data for my bedrock numbers and then the logic of the game progression builds the final outcome from there.
OK, fine. 
I’ve started looking into this.
The first thing I’ve done is to break out the cost range of assassination and spying hijinks into a table, then correlating the range of those costs against NPCs of various level’s income and assets - for example, the assassination of a 9th level NPC is a month’s wages for a level 8 person (as determined by the XP from Domain/Merchantile table). It’s around all the assets of a level 3 NPC however (by the ACKS Secret Ratio as applied).
So, in a broad brush, I know who can afford the services.
Second, I’ve constructed a table showing NPC population by level by market class, utilizing the Demographics of Leveled Characters table.
So, I know how many people can afford those services.
What I’m looking for now is, I think, historical or contemporary crime rates, for blackmail/extortion, espionage, and contract killings.
I take that data and apply it to my different population groups and I would then have some clue as to how many people per month want to hire an assassin based on real life, which then can be DND-ized into something probably much more violent.
Ignoring for the moment the type and number of watch lists this searching will put me on 
I’d welcome any suggestions about where I’ve started off wrong, or where to look offhand for more historical records, though I expect crime statistics weren’t much of a thing before the modern eras.
this might merit its own thread, you’re already getting pretty narrow boxes.
This definitely seems like the best place to start. it would be interesting to see how these things stack up alongside relative success rates. There may be fewer pepople who can afford higher level assassinations, but the higher level thief-type is also more likely to succeed.
There also seems to be a fundamental problem already highlighted: a level 8 can afford to hire an assassin for someone higher level than them once per month, ostensibly at little to no risk to them. That would suggest that people should be dying left and right, surely every level 8 has some level 9s they would like to see dead, or if not there must be plenty of level 7s, 6s, and 5s they fear trouble from.
I suspect that the actual net pay earned by the crime boss and the perpetrator must be only a percentage of the actual cost of accomplishing an assassination.
Yea, I’ll start a new thread here soon, perhaps once I’ve gotten something on crime rates settled.
So, to speak to what you said, a level 4 character can, if all his or her assets are liquefiable, finance the assassination of a 14th level character.
It’s the daily income of a level 13 character, if they’re operating above their GP threshold.
I’d expect in the end that yes, the profit from the venture is not necessarily what it costs to implement.
I bet there’s some sort of acceptable ratio between the value of the target in assets and how difficult it would be to make the attempt, perhaps with some sort of cost/time metric - I can kill the World Emperor for you, but it’ll take 15 years to get a girl in as his primary concubine - and you can’t pay enough to make it happen in 6 months.
(or, alternatively, under a market-based approach assassination just isn’t common enough to be a hijink - that’s certainly a possibility)
Well, in Alex’s defense, the book specifically calls out plot-based assassinations (which a world-emperor would always be by definition) as being outside the purview of the assassination hijink. An assassination hijink is something rote or mundane, not worth excessive detail regarding who gets offed, they’re somebody unimportant who made one too many mistakes in front of rich people.
This is true.
I wasn’t painting the current system as broken with that example, more that my own tables provided the image of hundreds of assassination attempts per day on any given 14th level character (visualizing some sort of Clouseau/Cato situation) which is amusing.
To turn that around, the assassination of a 1st level character is the daily income rate of a 10th level character, so that’s fine as long as you don’t have 6 fingers on your right hand and that weaponsmith has a particularly plucky son.
Above in this thread I briefly modeled some transactional data for cities and concluded that demand for goods in a city was about 80gp per family. I then did some further estimates that if all trade was done in the city, total transaction value might be 250gp per family.
With further refinement (via spreadsheets too complex to share here):
- the demand for goods is 75gp per family
- the total value of all market transactions is 100gp per family*
- The absolute maximum number of merchants available is 36 times the average number listed on the Markets and Merchants table.
*There are more transactions than the demand for goods from just the urban families because some transactions occurring are by peasants from the surrounding area, church leaders from tithes, etc.
This leads to the following maximum monthly volume:
Class I: 30 Large Ships, 90 Small Ships, or 45 30-horse Caravans
Class II: 7-8 Large Ships, 22 Small Ships, or 11 30-horse Caravans
Class III: 3-4 Large Ships, 11 Small Ships, or 6 30-horse Caravans
Class IV: 1 Large Ship, 3 Small Ships, or 2 20-horse Caravans
Class V: 1 Small Ship or 2 10-horse Caravans
Class VI: 1 10-horse Caravan
I probably need to create an NPC specialist class called an “agent” “broker” or “procurator” who can conduct dealings on behalf of PCs so that PCs who end up owning, e.g., 8 Large Ships can use brokers to handle their trade.
I may be off base here, but one of the things I’m noticing in some of the posts about Thieves Guilds dominating domain economics is the lack of direct, tailored opposition. What I mean by that is that when a PC builds a domain, there are hexes that have to be cleared out, and the Judge can work out how to model that from the existing rules. While you can work out what kind of Thieves Guild and/or local law enforcement is present to oppose Thief-type PCs, perhaps some guidelines for this would be a great addition to Lairs and Encounters, similar to the large-scale hex-clearing guidelines that are supposed to appear in that book…
There is that table with guild boss level by town size, which could be a starting point. Might be useful to further expand expected holdings of religious organizations for when the cleric starts trying to poach followers in a domain with an existing temple as well.
Yeah, that’s exactly what I’m thinking of. And you are correct: it’s not as if you can’t deduce the answer from the rules as they currently stand, it’d just be handy to have a suggested procedure and any shortcuts laid out.