Distributive Justice

What it is

The idea is older than marketing. In social exchange theory, people judge an exchange by whether what they receive is in proportion to what they put in, and they feel the gap when it is not. Service researchers borrowed that judgement and pointed it at complaints.

The borrowing that stuck came from Stephen Tax, Stephen Brown and Murali Chandrashekaran in the Journal of Marketing in 1998. Using justice theory, they showed that customers evaluate a complaint incident on three things: the outcomes they receive, the procedures used to arrive at those outcomes, and the nature of the interpersonal treatment along the way. The first of those is distributive justice. The other two are procedural justice and interactional justice.

What counts as a fair outcome is not only the amount. Amy Smith, Ruth Bolton and Janet Wagner built their 1999 model of failure and recovery on resource exchange, mental accounting and prospect theory, and found that restaurant and hotel customers preferred recovery resources that matched the type of failure, in amounts commensurate with its magnitude. Fairness of outcome, on that account, has a kind as well as a size.

In effect

The clearest evidence on what distributive justice does comes from Katja Gelbrich and Holger Roschk's 2011 meta-analysis of complaint handling in the Journal of Service Research. They tested a path running from what the firm does, compensation, employee behaviour and procedures, through the three kinds of justice, to satisfaction and then to loyalty and word of mouth. The model held. The surprise was where distributive justice sat in it. The common view that outcome fairness is the main driver of recovery was true only for satisfaction with the single transaction, which in turn fed positive word of mouth. Cumulative satisfaction, the main antecedent of loyalty, depended slightly more on interactional justice than on distributive justice. Outcome fairness mattered most when the complainants studied were students.

The size of the outcome has been tested directly. Gelbrich, Jana Gäthke and Yany Grégoire ran two experiments in 2015 with eleven compensation levels, from nothing to twice the value of the loss. Where customers had rejected the flawed service, satisfaction rose most steeply somewhere between about 60 and 120 per cent of the loss. Where customers had accepted the flawed service and lived with it, the first money offered bought the most satisfaction. Neither curve rewards generosity far beyond the loss.

The kind of outcome matters as well. Roschk and Gelbrich's 2014 meta-analysis and experiment found the recovery effect was generally strongest when compensation was a resource similar to the failure: immediate money for a monetary loss, an exchange for a flawed product, reperformance for a failed service. Immediate money generally did more than delayed money.

And the outcome interacts with the rest. In Jochen Wirtz and Anna Mattila's 2004 experiment, compensation did not raise satisfaction when the response was already immediate and came with an apology, and it failed to lessen dissatisfaction with a delayed response that had none. It helped only in mixed cases. Their phrase was that compensation is a poor substitute for a good recovery process.

What it does not say

It does not say compensation is unimportant. For satisfaction with the incident it is the main driver, and a customer who has lost money and is not repaid has a grievance that no apology covers.

It does not say there is a correct percentage. The 60 to 120 per cent band comes from two experiments on one kind of flawed service, and the authors' own point is that the shape of the curve depends on whether the customer rejected or accepted what went wrong.

It does not say the three kinds of justice are independent. In Wirtz and Mattila's design the three interacted, and the effect of the outcome depended on what the procedure and the treatment had already done.

And most of the evidence comes from scenarios and experiments in which people respond to a described or staged failure rather than their own. Tax and colleagues' study of real complainants is the main exception, and its finding was that a majority were dissatisfied with how they had been handled.

---