.
1. Warm glow feeling, altruism, paternalism, moral sentiment.

I'm stuck with this paper. No significant progress after almost two weeks struggling with incorporating warm glow effect proxy into my model. I hypothesized that some part of households' willingness-to-pay for harbor cleanup in Waukegan , IL are in fact due to some warm glow feeling, i.e. not purely out of concern about environment as public good.

Problem is, we didn't specifically address direct questions for this in the survey. Nunes and Schokkaert (JEEM, 2003), for example, have attitudinal questions like "Our family admires the individuals who, on voluntary basis, participate in collecting donations for national programs for social aid and solidarity", or "There are some funding campaigns to which my family and I feel very close and therefore we do not hesitate to contribute a donation", etc. These questions obviously relate to detecting warm glow feeling and can be used for that purpose accordingly.

Unfortunately, all we have in our survey instrument is a set of attitudinal questions that do not directly address this issue. We asked respondents their Likert-type agreement on statements such as "Harbor area is safe", or "Harbor area is economically enhancing", etc. In addition to that, we have perception questions, also in Likert-scale, namely "Harbor pollution is not-at-all-important; or somewhat-important; or very-important". Hm, now I have a vague proposition: "If a household reveals a higher-than-average willingness-to-pay for harbor cleanup, while in fact it stated that harbor pollution is not-at-all-important, then warm glow effect is present". OK, how to model this?

Another complication is due to this whole debate on the definition of "warm glow effect". An article forthcoming in J. Economic Psychology (2003) by Elias Khalil actually addresses this. But it's a little too broad and descriptive without formal model proof (well, I guess that's what I am supposed to pursue). He divides theories of altruism into rationalistic and normative. The first one is further decomposed into egoistic, egocentric, and altercentric perspective. The second one into Kantian, socialization, and warm glow. (There you go...)

Further, summarizing Khalil's rationalistic theories of altruism: "egoistic" altruism (e.g. Axelrod) says that "altruistic assistance would be offered if one expects future benefit". "Egocentric" altruism (e.g. Becker's rotten kid theorem) argues that "donor's utility function includes [that of] potential recipients". While "altercentric" altruism (e.g. Frank, Simon) views benefactor's action as "stemming from a personality trait that arises from artificial selection". Phoohhh...

Enter normative theories. Kantian perspective (e.g. Etzioni) resembles the altercentric approach. However, Kantian cannot distinguish between altruism and "honesty": humans should not deceive others. Socialization perspective (e.g. Mead), on the other hand, believes that agents tend to act in particular ways to gain "approval, respect, admiration, and prestige" accorded by some peer group. Now the third one: "Warm Glow" story. Since this is really my main interest, let's dedicate a separate paragraph:

Arguably, the term "warm glow" was coined by Jim Andreoni (e.g. JPE, 1989). He defines "impure" altruistic action as the act that is partially motivated by the "warm glow" and purely by the concern over the beneficiary's welfare. According to Khalil, Andreoni uses the notion of "warm glow" to explain the puzzle of "why altruists do not greatly free ride". Khalil claims that Andreoni's "warm glow" is just a "by-product of doing the right thing". -- Now, my proposistion above seems to have a rationale...

But then, this warm-glow business has been so controversial in the field of environmental economics. Started by Kahneman and Knetsch's paper criticizing contingent valuation, in JEEM (1992) that says that what people purchase in regards with nonuse values is "moral satisfaction". This paper was instantly attacked by Smith (same journal, same issue) and Anderson (same journal, next issue).

Then there goes the classic economists debate over Exxon-Valdez incident (1992, see links Oct 6 entry below). The famous book edited by Hausman documented the somewhat-mean-spirited debate. Hausman, Diamond, Desvouges, Boyle, McFadden, Milgrom, Ken Arrow, etc. versus Carson, Rowe, Kerry Smith, Rich Bishop, Hoehn, Kolstad, Bockstael, etc. [See also the 1994 heated symposium in J. Econ Perspectives between Portney (neutral), Hanemann (pro), and Hausman and Diamond (con) -- available from JSTOR). Unfortunately, only one article in the Hausman book (the one by Paul Migrom) specifically addresses this issue of warm glow. Reading the discussion, it seems to me that even the panelists and discussants have different definitions of warm glow in willingness-to-pay for nonuse values. And who am I to define it myself?

2. Having done literature review on this warm-glow-effect-business, I conclude that there has been zero attempt to analyze warm-glow share in willingness-to-pay in the context of choice experiment combined with hedonic approach. Provided that I come up with a good, convincing argument in the paper, this work will be the first of it's kind. Admittedly, Kurayama from Hokkaido University has pioneered warm-glow test in a choice experiment (e.g. his working paper, 1998). However, as it turns out, he doesn't need to use hedonic approach, as he analyzes the willingness-to-pay for forest preservation and one of the attributes in his choice sets is the amount of tax or direct contribution to the preservation effort. His theory part is standard and (therefore, should be) convincing. Using Hicksian compensating surplus (CS) concept, he goes on distinguishing between total CS, ex ante CS , and ex post CS. He claims that the pure value of environmental change is the difference between the total CS of environmental quality change and the ex post CS of warm glow. However, when it comes to empirics, I find it much less convincing. In order to capture the warm glow effect, he just adds a dummy in utility function, defined as 1 if tax is positive and 0 otherwise. Why so ad hoc?

Our study on the other hand doesn't have this "privilege" of payment vehicle that can serve as direct anchor for willingness-to-pay for change at stake. We in fact have to do some "magic" (that is, hedonic approach) to squeeze environmentally-related value out of housing buying price. This is what will make this paper contributive, I hope.

2. Reflection: Becker, 1993. Nobel Lecture: The Economic Way of Looking at Behavior. JPE 101(3): 385-409.
Quoted: [I]ndividuals maximize welfare as they conceive it, whether they be selfish, altruistic, loyal, spiteful, or masochistic.