You may have noted from Table 1A that the number of tokens awarded (or taken away) for each target behavior varies, as does the number of tokens required to purchase each backup reinforcer in Table 1B. Both of these factors are balanced in order to determine the amount of behavior per backup reinforcer, or the price.
With regards to the number of tokens associated with each target behavior, there are several factors to consider (Sarafino, 1996). First, there is the law of supply and demand. Some target behaviors are less attractive than others. Maybe they are more strenuous and time-consuming (e.g., cleaning one's room versus turning out lights), more tedious (e.g., doing dishes versus watching news on TV), or more anxiety-producing (e.g., performing homework versus reading the newspaper). More tokens should be given for less attractive target behaviors. Second, when a token economy involves a group of individuals, the number should be adjusted based on the abilities of each individual so as to equalize the potential for earning backup reinforcers. Joe, who has always done well in school when he applied himself may be awarded 1000 points for earning an A grade in any of his classes; Dennis, another group home resident who tries hard but has never done well in school, may be awarded an equal number of points for earning a C grade in those same classes.
The number of tokens lost for each undesired target behavior should be determined by the severity of the problem. As you can see in Table 1A, stealing, lying and cheating are deemed to be very serious offenses at Achievement Place. However, this needs to be weighed with the fact that taking away too many tokens too often puts the individual in a position in which he or she has nothing left to lose. Thus, the cost of a fine should also be determined by the relative number of tokens that can potentially be acquired and the price of the backup reinforcers (Miltenberger, 1997). With fining (also called response cost), one hopes to lessen the opportunity to purchase backup reinforcers, but not to negate that opportunity altogether. Participants must exchange tokens for backup reinforcers at least once in a while in order to establish the tokens as conditioned reinforcers. Otherwise, the tokens will become meaningless - their acquisition will not reinforce desirable target behaviors and their loss will not punish undesirable target behaviors. For these reasons, it has been recommended that if a token economy is to include a response cost component, it should be introduced after the token economy has been in place for a while (Miltenberger, 1997).
With regards to the number of tokens required to buy each backup reinforcer, there are some additional considerations (Sarafino, 1996). First, the token economy price should reflect the actual monetary cost of that item, if for no other reason than to avoid a collapse in the economy! Second, the laws of supply and demand should apply. If functioning in the role of the manager for the day is an activity desired by many of the residents at Achievement Place, and only one of them can hold this "prestigious" position at a time, then a higher cost can be justified for that privilege relative to one with an unlimited supply (e.g., watching TV). Third, the therapeutic value of a backup reinforcer should be inversely related to its cost. For example, access to tools in the workshop for juveniles at Achievement Place were included with "Basics" and cost less than watching TV (see Table 1B), probably because residents have more to gain by working with tools (e.g., developing a skill that may ultimately secure them employment outside of the group home).