Before all else in this post: I am not an Economics major. I am, moreover, aware that I am not an economics major. I do know a few things about games, though. (Commentary by actual economics majors is welcome, though.)
I've been reading the Valve economics blog, and it got me thinking. First of all, the Valve blog is mostly concerned with player-to-player trades; interesting, but not the kind of economics that roleplayers deal with the most. Of greater interest to LARPers, tabletop gamers, MMO players, et al., is the economy of interacting with the game environment. (I'm ignoring for a moment the huge interest in and valid study of MMO auction houses.) I was originally going to discuss LARP economies in detail in this post, but I've cut those to save for a later post - staying on-topic is an issue.
Roleplaying games that worry about having an economy in the first place are usually interested in simulation, to a greater or lesser degree. Most games stop short of getting into the nitty-gritty of bills and maintenance, either because those things aren't fun, or because PCs are rootless wanderers who spend most of their nights under the stars or in a dungeon. As I mentioned in this post, Pendragon and AD&D 2e are notable for having definite, if abstracted, maintenance rules.
These two games each fix some but not all of the notable problems with upkeep costs: Pendragon includes a very clear phase of play in which to pay for upkeep, and grants mechanical benefits (both tangible and highly desirable in long-term play) to paying the optional larger upkeep costs. Having a definite phase of play for upkeep is useful because it keeps it from getting glossed over (with the downside that it can take half an hour, if everyone is on-task, to process upkeep and dice rolls for six or seven players). Having optional larger upkeep costs and mechanical benefits is good because it phrases those costs as a benefit to the player; considering that the player earns Glory from conspicuous consumption and mitigates some of the system's absolute squalor, I know that I tended to pay Superlative-level maintenance (a painful £12 per year) as often as I could afford it.
2e includes a scaling cost, based on player level, for the highest upkeep options. This is good because treasure values tend to increase over the course of a game. It's obvious that they would do so, if you're thinking of the game as a fantasy; once you've achieved an impressive haul, the next thing you want to do is find an even bigger treasure. I am pretty sure explaining concepts of progress and reward won't be necessary here. The problem with them, of course, is that an inflating reward is kind of antithetical to a functioning game economy. The scaling upkeep cost, then, means that players have to turn over higher profits at higher levels just to break even. Does this make any sense in the setting? Not without some substantial help from the DM. This combines with what I think is an expectation that players of 9th level and above will settle down, manage baronies, and adventure only intermittently; in that light, some of the Monty-Haul-ish treasure hoards that the DMG fully prepared to churn out make a lot more sense. But, well, we didn't play that way, and the DMG was much too vague in stating these expectations.
2e and 3.x/4e have completely different assumptions about what players want to buy with their epic piles of gold, as well. 2e belongs to the era of its predecessor editions, as well as Pendragon and ACKS, in assuming that players will want to transition into domain-style play. I can't help but see this as an exact parallel of players in MMOs reaching the endgame and deciding that player housing is the new style of gameplay that they want to open up. It just happens that the 2e designers (and Gygax before them) thought 9th level was about the right time for that shift. There's still room for traditional dungeon-crawling adventure, of course, but domain-style play is hard-coded into the class abilities of many character classes. 3e and later editions assume that high-level mathom-hoards go to buying or crafting still better items - call it a gear treadmill if you like, but it's a cycle of play that has gotten its hooks into plenty of players over the years.
I think every DM sooner or later gets into the situation where the party stays in an inn and the innkeeper charges what the non-adventuring world regards as an unbelievable amount of money (one gold coin, let's say) because the players are sitting on such unthinkable piles of wealth that they don't really remember that copper and silver coins are an option, or because it's not worth anyone's time to note an expense of less than one gold coin. There's an argument to be made that innkeepers are not above gouging these obnoxious, wealthy adventuring types, or that the now-wealthy characters are seeking out more upscale residences than the roach motels they once occupied. The same thing has an overwhelming tendency to happen in LARPs, but it's only really noticeable in player-run bars. (Let me be clear: I'm not criticizing - there are a lot of challenges and demands in running any in-play location, and none of my PCs have ever been directly involved with them.) The players are charging for drinks what it seems reasonable to charge, based on how much money the average player can pay without having to think about it too hard. My recollection from Shattered Isles, King's Gate, and Eclipse is that this will usually increase by a factor of 10 to 20 over the course of campaign - of course this is no more than a recollection of anecdotes, not data. Player perception of money both causes and results from this change over time.
How Value Enters the Economy
I suspect that this is the area in which gaming economies completely leave real economics behind. If anything, though, a designed game economy resembles a discussion of federal monetary policy: money flows forth from the void, and those in charge must decide how much and how easy it will be to acquire in order to avoid runaway inflation. At the same time, they need to keep things interesting. Unlike the Fed, though, money leaving players' hands returns to the void, losing all sense of existence beyond pure fiat. If merchants in the real world were motivated not by profit, but by making sure that money would continue to motivate the customer, the rest of the metaphor might work. Anyway, since we've talked a bit about the passive, frequently zero-perceived-benefit ways the money leaves the economy, let's talk about how the money got into player hands in the first place.
In D&D-style tabletop games, PCs gain wealth primarily through looting enemies, capturing treasure hoards, and getting paid by NPCs for completing quests. The first of these is generally small, and as long as the amount of coin graduates from silver to gold at about the same time that players stop caring about the existence of silver, it doesn't need to become a significant source of coinage. On the other hand, players probably will expect slain enemies to use magic items that the PCs can now claim for their own, as the power level of the campaign enters the middle levels, much more so when fighting enemies who parallel them in abilities and nature (or even just basic physical shape, such as giants) and less so when fighting beasts, constructs, elementals, oozes, and so on. Players generally expect treasure hoards and quest-reward payments to top their last major haul, as the challenges they have just defeated are probably greater than they have previously faced, and the order of the adventuring universe suggests that reward should follow risk. I did once run a campaign in which, thanks to varying up the denominations used, I handed out a huge cash sum in about the fifth session, and smaller overall values but impressively large numbers of coins later on. Maybe the players noticed and maybe they didn't, but they never mentioned dissatisfaction with the flow of coin. Oh, and it's nominally possible to earn a bit of cash on the side from the Craft, Profession, Perform and Tumble skills; even at very high skill levels Craft and Profession earn next to nothing, while Perform and Tumble are reasonably lucrative for busking - good for some pocket change, at least, but never intended to be a major source of income unless the campaign has frequent month- or season-long downtimes.
Pendragon is an interesting special case for its approach to income as well. A knight's three primary sources of income are loot taken after a battle, ransoms of enemy knights left alive, and income from the knight's manor(s). Loot won in battle is entirely unreliable, as the Saxons occasionally spend a few years licking their wounds - especially after Mynydd Baddon. Ransoming enemy knights (which, fortunately, can include Saxons) is chancy - sometimes his family has no money, and chivalry demands that you pay for that enemy knight to crash on your couch until they raise the money, or the depredations of the British climate take their toll on him. When it works, though, ransoming knights is the best money out there... which is good, because capture is never certain (a crit could kill the enemy outright), it takes time out of a battle to secure a prisoner (when a knight could otherwise be earning more Glory), there's a limit to how many prisoners a knight can keep (if I recall correctly), and knights receive more Glory for a kill than a capture. Finally there's manor income; a knight with only one manor generally won't do any better than breaking even on the manor's expenses, but two or more manors might make enough money to cover the knight's upkeep. Assuming the weather doesn't crit, he has a decent steward (a wife, by preference), and bad fortune doesn't get in the way, that is.
In our experience of Pendragon, at least, there wasn't a lot of questing after big piles of silver or looking for the next big score. Treasure outlay does increase over the course of a single character's career: improvements made to a manor generally cause it to bring in more money, characters are more likely to gain additional manors than lose existing manors, Stewardship stats can only increase over time for any given character, and it becomes possible to capture or slay kings during and prior to Mynydd Baddon. Kings, predictably, are flush with wealth; I don't know if it is standard for Pendragon or our GM's convention that kings consistently have one piece of magical loot. Still, silver comes and goes, and times can still be tight for years on end if a knight's luck turns sour. (The problem with the game is that shaking off the effects of this bad luck and catching up with other players is just about impossible, and with so little systemic niche protection, the encouragement to compete is intense.)
Getting Money Out of the Economy
As I suggested above, GMs want to get money back out of player hands, for a number of reasons. The primary reason is continuing motivation. Now, it's true that for some players the number next to GP is just a way to keep score. For me, though, and I think for a majority, money is only worth what you can get for it. If I don't believe I'm going to be able to convert that vast pile of gold into an interesting object (whether it's a sword, hat, or castle with more ridiculous pluses than the one I had before), gold ceases to motivate me. One could reasonably ask why a player wants those things in turn; I think the answer comes down to making a definitive and compelling mark on the shared fiction, whether through defeating the most powerful monsters or getting everyone at the table to agree on imagining the player's castle as being totally awesome.
Anyway, you didn't come here for a discussion of why things motivate people. The other reason the GM wants to remove cash from the economy (a distant second place) is for verisimilitude, sustaining a certain feel in the game. It's hard to make the game feel anything other than over-the-top when players start to think about the poundage of the coins they're toting around and realize that if the encumbrance rules pay attention to coins, they might need some help... and that starts to change the desired image of their characters going about town, and so on. There are plenty of people who won't care about this aspect, of course. An arguable third is pure perception: even if you've spent large piles of money before, if you're just about skint now, even modest rewards seem more interesting, because it's easier to compare a new amount of money to your current finances than to what you remember spending in the past.
I'll offer the idea, then, that fantasy adventuring games along the lines of D&D (much less so with Pendragon) should work on adding two types of things to their item lists. First, more variety of consumable items, including broad support for characters built on consumables, such as alchemists. This is something D&D has never done all that well, in that acid and alchemist's fire have been underwhelming for some time now, and 3.x's tanglefoot bags only retain their utility because they have an effect even on a successful save, and that effect is not a damage value. Of course, to get anywhere with this, they'd probably also want to revamp the crafting rules, and I don't have a lot of hope for that. The other kind of item I'd like to see added is the long-term consumable - something that, once activated, lasts for anywhere from a week to a year, but will eventually need to be replaced or maintained.
I've suggested limited-duration enchantments before, but I want to make the case for them again. Now that 5e is going away from "assumed math" and extensive gearing just to keep up with the monsters (or so they say), I'd like to see permanent magic items become much rarer, while temporary magic items are a calculated decision that a character makes; if you don't think you need the magic in order to succeed, then save your money. The haul from your current task (along with other sources of income, such as funding by a holy order or whatever) is what puts you in position to pay for the next adventure, ideally with a tidy profit. The secondary benefit is that bad guys can be equipped with magic items without the setting being flooded with magic items priced at 1000-2000 gold each, as is the case in 3.x with amulets of natural armor and rings of protection. The costs to renew enchantments don't need to be as steep as those to place the enchantment in the first place - in fact it's great if they're not, so that players are less inclined to tailor every item to every mission, but still can do so if they like.
Getting players to invest, both financially and emotionally, in the game's organizations and edifices is excellent, though some thought needs to be given to mechanical dimensions to help them feel real and present. Collective resources like these have (quite rightly) become very popular in the game design of the past few years, and they're great money sinks as long as the DM is okay with the collective resource actually improving in a way that feels meaningful in the game.
Next time I come back to this topic, I'll try to focus on LARP economics, using the same areas of upkeep, income, and outlay.