Token economies can be time consuming and expensive. Gauging the target behaviors before implementing the program will indicate the extent of the problem and whether or not components of a token economy, or even the token economy itself, is justified. For example, if the before-treatment data indicate that undesirable behaviors are infrequent, then a token economy might focus solely on appropriate behaviors that earn tokens. Gauging the target behaviors during a token economy - and comparing this with the before-treatment levels - helps determine whether or not the target behaviors are improving. If so, the person might then move to the next stage of the program; if not, the program might need to be changed in some way. In Joe's token economy, the data would determine when he advanced from the daily to the weekly exchange of tokens for backup reinforcers. Basing this decision on subjective impressions may result in Joe moving forward before he was ready and ultimately failing because acquisition of the backup reinforcers became too temporally remote from earning the tokens.
A goal in many token economies is to help clients adjust to normal living conditions outside the setting in which the token economy is implemented. Unfortunately, as noted earlier, Joe's improvements achieved during his stay at Achievement Place are unlikely to maintain when he is transferred to his normal home environment. But is this surprising? He will probably be returned to the same situation that was responsible for his delinquency to begin with, one in which prosocial behaviors are ignored and antisocial behaviors are reinforced. One solution that has been proposed for juveniles such as Joe is an extended supportive environment: "The relatively permanent arrangement of socioenvironmental conditions to provide ongoing support" (Wolf et al., 1987, p. 348).
Martin & Pear (1998) note that "because social reinforcement, not tokens, prevails in the natural environment, a token economy should be designed so that social reinforcement gradually replaces token reinforcement" (p. 303). There are a variety of ways to do this. One might eliminate tokens gradually by requiring greater amounts of behavior to earn the same number of tokens, restricting the range of behaviors that earn tokens, and increasing the delay between the target behavior and token delivery. Or, one might decrease the value of the tokens by increasing the cost of backup reinforcers or by increasing the delay between acquiring tokens and redeeming them for backup reinforcers. For an extended discussion of this topic, see Kazdin (1977, Chapter 7).
Note, however, that promoting generality need not be a goal of a token economy. For example, a token economy was instituted to improve worker safety at two open-pit mines (Fox, Hopkins, & Anger, 1987). Trading stamps were earned for various safety-related behaviors (e.g., working without lost-time injuries), and could be exchanged at neighboring redemption stores that carried a selection of several thousand items. The resultant reductions in injury costs (e.g., number of days lost from work because of injuries) greatly exceeded the costs of operating the token economy, and it remained in effect for over 10 years. Practically, there was no reason to tamper with the prevailing contingencies. In fact, the company that operated one of the mines started two new strip mines, both of which adopted an identical token economy.
Critics have described token economies as "demeaning," especially to adults. To this criticism Spiegler & Guevremont (1998) reply:
"The argument is that token economies are appropriate for children, but not for adults, who are supposed to be 'above' receiving tokens for behaving appropriately. Interestingly, the same people who think that token reinforcement is inappropriate for adults often forget that they participate in a large-scale token economy every time they make a monetary transaction... In fact, the evidence suggests that token economies enhance clients' dignity by increasing their self-esteem, self-respect, pride, and sense of worth... [C]onsider the comment of a highly intelligent, 40-year old trainee at the Community Training Center. The man approached the program director and, with obvious pride in what he had accomplished in his classes, said, `Doctor, I earned 120 credits today.' These words sound very much like a sales executive telling her husband at dinner, `I earned a bonus today for closing a deal,' or a college student telling his roommate, `I got an A on the paper I worked so hard on.'" (pp.189-190).
What would Joe ultimately think about the token economy at Achievement Place? If past research with his peers at that group home is any indication, he will give a satisfaction rating of about 6.5 out of 7 points; ratings given by juveniles in group homes using other approaches average only 5 points (Kirigin et al., 1982).