status quo bias in the measurement of value of service

18
Resources and Energy 12 (1990) 197-214. North-Holland STATUS QUO BIAS IN THE MEASUREMENT OF VALUE OF SERVICE* Raymond S. HARTMAN University of California, Berkeley, CA 94720, USA Law and Economics Consulting Group, Berkeley, CA 94710, USA Michael J. DOANE and Chi-Keung WOO Analysis Group, Inc., Sausalito, CA 94965, USA Received August 1989, final version received April 1990 Notions of customer value of service have become increasingly important in utility resource planning, capacity expansion and rate making. In order to design and implement value of service policies, utilities have come to rely upon a variety of customer surveys. Unfortunately, theoretical and empirical analysis suggests that responses to such surveys may be seriously distorted by ‘status quo’ effects. Using both regression and choice-theoretic frameworks, we empirically investigate whether status quo effects arise in a contingent valuation survey addressing the value of service reliability for residential customers. We find substantial status quo effects, which must be explicitly understood and addressed. 1. Introduction and overview The notion of ‘value of service’ has received increasing attention in a variety of contexts. In the context of gas and/or electric utility marketing efforts, value of service concepts have been used to address customer preferences for the attributes of a particular energy source relative to other fuels and/or cogeneration technologies. In the context of utility rate making, value of service measures have been used to analyze customer demand for and acceptance of priority service. In this paper, we focus upon specific measurement issues that arise when attempting to measure the value of priority service. In general, priority service is a contractual mechanism for allocating scarce supplies of a service or product. A seller provides priority service by offering a menu of contingent forward contracts which are differentiated by the priority with which the service is made available. Based on the perceived importance of *The empirical work reported in this paper was funded by the Pacific Gas and Electric Company (PG&E). The views expressed are those of the authors and do not necessarily represent the opinions of PC&E. The authors gratefully acknowledge the comments of William Schulze and an anonymous referee and the research assistance of Renee Rushnawitz. 0165-0572/90/$3.50 0 1990, Elsevier Science Publishers B.V. (North-Holland)

Upload: raymond-s-hartman

Post on 30-Aug-2016

213 views

Category:

Documents


1 download

TRANSCRIPT

Page 1: Status quo bias in the measurement of value of service

Resources and Energy 12 (1990) 197-214. North-Holland

STATUS QUO BIAS IN THE MEASUREMENT OF VALUE OF SERVICE*

Raymond S. HARTMAN

University of California, Berkeley, CA 94720, USA

Law and Economics Consulting Group, Berkeley, CA 94710, USA

Michael J. DOANE and Chi-Keung WOO

Analysis Group, Inc., Sausalito, CA 94965, USA

Received August 1989, final version received April 1990

Notions of customer value of service have become increasingly important in utility resource planning, capacity expansion and rate making. In order to design and implement value of service policies, utilities have come to rely upon a variety of customer surveys. Unfortunately, theoretical and empirical analysis suggests that responses to such surveys may be seriously distorted by ‘status quo’ effects. Using both regression and choice-theoretic frameworks, we empirically investigate whether status quo effects arise in a contingent valuation survey addressing the value of service reliability for residential customers. We find substantial status quo effects, which must be explicitly understood and addressed.

1. Introduction and overview

The notion of ‘value of service’ has received increasing attention in a variety of contexts. In the context of gas and/or electric utility marketing efforts, value of service concepts have been used to address customer preferences for the attributes of a particular energy source relative to other fuels and/or cogeneration technologies. In the context of utility rate making, value of service measures have been used to analyze customer demand for and acceptance of priority service.

In this paper, we focus upon specific measurement issues that arise when attempting to measure the value of priority service. In general, priority service is a contractual mechanism for allocating scarce supplies of a service or product. A seller provides priority service by offering a menu of contingent forward contracts which are differentiated by the priority with which the service is made available. Based on the perceived importance of

*The empirical work reported in this paper was funded by the Pacific Gas and Electric Company (PG&E). The views expressed are those of the authors and do not necessarily represent the opinions of PC&E. The authors gratefully acknowledge the comments of William Schulze and an anonymous referee and the research assistance of Renee Rushnawitz.

0165-0572/90/$3.50 0 1990, Elsevier Science Publishers B.V. (North-Holland)

Page 2: Status quo bias in the measurement of value of service

198 R.S. ~ariman et al., Status quo bias

the service, each customer in the market selects a specific contract with concomitant service priority. The seller rations the service, in the face of supply constraints, on the basis of the contracted priorities.

For electric and gas utilities, priority service takes the form of service reliability. In particular, utilities attempt to design reliability-di~erentiated rate options for heterogeneous customer groups,’ Customers with a greater willingness to pay (WTP) for reliable service receive it with a greater probability by selecting a service contract with a higher priority of service. In the event of a capacity shortage, excess demand will be efficiently rationed by allocating scarce supplies to those customers who have reveaied a higher value of service by their selection of a higher priority contract.’

In order to identify appropriate levels of system reliability and to design and implement specific reliability-differentiated service contracts, utilities require accurate information on the value of service (willingness to pay) for heterogeneous customer groups. In developing such information, utilities have come to rely upon survey information, and because reIiability- differentiated contracts are essentially new services, contingent valuation methods have been frequently exploited in the surveys.

Wnfortunately, recent theoretical and empirical analysis suggests that such survey information may be less useful than originally thought. The major thrust of this analysis is that a survey response is contaminated by the respondent’s current situation (his/her status quo).3 For example, contingent valuation studies have identi~ed a status quo effect in the form of asym- metric valuations of losses and gains from the status quo. In particular, willingness to accept (WTA) measures of value have often exceeded willing- ness to pay (WTP) measures by an order of magnitude of three to one.4

In this paper, we empirically investigate whether such status quo contami- nation is found in customer valuation of service reliability. We base our analysis on the results of a contingent valuation survey of residential customers in the Pacific Gas and Electric, Co. (PG&E) service territory. In the survey, each customer was asked to provide WTA and WTP measures of

‘For example the California Public Utilities Commission explicitly recognizes the need for unbundling traditional energy services in Decision 86-012-010, December, 1986. In this Decision the state utilities are allowed to negotiate rates with Iarge natural gas users that reflect a separate rate element for their desired priority of service.

‘Chao and Wilson (1987) demonstrate that implementation of reliabii~ty-di~erentiated priority service offers substantial gains in allocative efftciency in the industry. They demonstrate that priority service contracts can be used to approximate spot pricing in electricity markets. For related developments, see Tschirhart and Jen (1979), Oren et al. (1986), Bohn et al. j1984), Doane, Hartman and Woo (198Xa, b), Hartman and Woo (1988), Muoasinghe (1979), and Munasinghe and Gellerson (1979).

3Kahneman and Tversky (1979) were the first to formalize the importance of a status quo effect. Machina (1987) and Hartman, Doane and Woo (1990) summarize more recent literature.

%ee Bishop and Heberlein (1979), Rrookshire and Coursey (1987), Cummings, Brookshire and Schulze (1986), Knetsch and Sinden (1984), and Rowe, d’Arge and Brookshire (1980).

Page 3: Status quo bias in the measurement of value of service

R.S. Hartman et al., Status quo bias 199

value for service reliability. Reliability was described by the presence or absence of service disruptions (i.e., power outages) with varying attributes (e.g., season, time-of-day, duration and the extent of advanced notice). The WTP measure represents the dollar amount customers would be willing to pay to avoid an additional service disruption; the WTA measure represents the dollar amount they would be willing to accept to incur an additional disruption.

In addition to these standard contingent valuation questions, each cus- tomer was presented with a menu of six alternative rate options designed to reflect varying levels of service reliability. One option characterized the reliability experience and current service contract of the customer, i.e., his/her status quo. The other live contracts offered options with varying bill discounts and altered levels of service reliability. From this menu, the customer was asked to identify his/her preferred option.

Using these data, we estimate and compare respondents’ WTA and WTP measures for several levels of service reliability. Furthermore, we examine and analyze respondents choices among the reliability levels offered by the six alternative rate options. Using a choice-theoretic framework, we quantify the determinants of these choices and calculate the compensating variations required to make customers indifferent between particular options. These analytic methods and the data are discussed in section 2. The estimated models are presented in section 3.

Section 4 discusses our conclusions. To summarize, we find that status quo effects consistently influence consumer valuations of electric service reliability. The WTA and WTP estimates for our sample customers differ by an order of magnitude of four to one, larger than expected from any reasonable income effects.5 Customer choice of reliability-differentiated service contract corroborates the importance of status quo effects. The resulting implications for utility policy are discussed.

2. The analysis

We use the analytic framework described in Hartman, Doane and Woo (1990) and summarized in fig. 1. I, represents an indifference curve reflecting consumer tradeoffs between service reliability (measured as decreased outage hours) and all other goods and services (measured as income, net of electricity expenditures). We assume that the income effect of marginal

‘These results corroborate the growing contingent valuation literature on consumer ‘irrationa- lity’ and the importance of the status quo [Brookshire and Coursey (1987), Coursey, Hovis and Schulze (1987), Machina (1987), Kahneman and Tversky (1979), and Samuelson and Zeckhauser (1986)].

Page 4: Status quo bias in the measurement of value of service

200 R.S. Hartman et al., Status quo bias

Other Goods (Income)

WTP,

Reliability

Fig. 1. Tradeoffs between service reliability and all other goods

reliability changes is small.‘j A customer is assumed to have an intitial service contract (a1 - characterized by monthly bill and number of outage hours) reflecting his/her current reliability. The actual service contracts are more complicated, as described later.

Along I,, the willingness to accept (WZ’A,) a marginal decrease in reliability (by an outage hour) is approximately equal to the willingness to pay (WTP,) for a marginal increase in reliability (by one outage hour). The slope of I, at a, is d, reflecting the relative price (or value) of reliability. If b, and c1 represent alternative service contracts (alternative reliability levels and alternative monthly bills), only slight changes in the price of reliability (slope d) will be required to stimulate the customer to switch to an alternative reliability regime (b, or ci).

When status quo effects are important, however, the indifference curve will be kinked at the status quo. I, (or I’) obtains, and the willingness to accept a one unit decrease in reliability (WTA, along Z2) is substantially larger than the willingness to pay for a unit of increased reliability (WTP,). At the same time, the change in the price (value) of reliability (slope d) required to induce the customer to switch from reliability regime a, to either b2 or c2 is now much greater or much less (respectively) than along I,. If the hypothesized

6For our sample, electricity expenditure represents a small portion of a household’s annual budget, approximately 1.3% (%400/%30,000).

Page 5: Status quo bias in the measurement of value of service

R.S. Hartman et al., Status quo bias 201

kink in utility obtains, disparate WTA and WTP valuations will occur, equivocally suggesting quite different values of service.

In order to investigate the divergence between WTA and WTP measures, a standard contingent valuation survey was used. The survey characterized the status quo (al) and socioeconomic attributes of the respondents. The customers were asked to state the dollar amount they would be willing to pay (WTP) to avoid a service outage as well as the amount by which they must be compensated to be willing to accept (WTA) the outage.’ WTA and WTP measures were obtained for the nine outage scenarios described in table 1.

The customers were then stratified into two groups on the basis of their current service reliability. One group had experienced, on average, approxi- mately three outages per year while the second group had experienced approximately 15 outages per year. The customers in each group were then presented with a menu of six rate options designed to reflect alternative reliability levels as well as their existing service reliability (i.e., status quo). Each customer was asked to rank the options in order of desirability. The reliability contracts for the two groups of customers are presented in table 2.

We analyze these data in two ways. First, we examine the WTA and WTP estimates for each outage scenario, focusing on sample means and distribu- tions. We find that the average WTA is substantially higher than the WTP for each scenario, supporting the hypothesized kink at the status quo (al). Because we also find that a substantial number of survey responses for WTP and WTA are truncated at zero (see table I), we also perform a two-stage Tobit analysis to eliminate any truncation bias that might affect these

‘To ensure reliable WTA estimates, respondents were asked to identify the actions normally taken to mitigate the effect of an outage and to consider the cost of such actions when estimating their WTA. To facilitate this process, a list of mitigating factors was provided and included such actions as the use of candles for light, eating out, the use of a backup generator, etc. We assume consumers would be willing to accept an amount covering these costs. To estimate WTP, each respondent was asked to state the amount he or she would be willing to pay for a backup generator service. This backup service was described as one that would handle all of his/her electrical needs during the outage. As discussed in Schulze, d’Arge and Brookshire (1981) and in Freeman (1982), our WTA and WTP measures may be subject to strategic response bias. However, previous work has found such biases to be small [see Mitchell and Carson (1981) Rowe, d’Arge and Brookshire (1980), Scherr and Babb (1975) Smith (1979)]. An anonymous referee has correctly suggested that strategic behavior will reinforce prospect theory behavior in the survey responses, if the survey sought to identify a single level of reliability as a public good for the service territory. Specifically, if a customer treats the survey as a single shot Prisoners’ Dilemma game played with his neighbors to contribute (be compensated) to (from) the utility for a unique level of system reliability, he will strategically lower (raise) his bid (e.g., WTP< WTA). Such strategic behavior does not arise in our analysis. A unique level of system reliability is not the subject of the survey. Different outage experiences occur for different customers. Different reliability levels and contracts are made available to all customers. Indeed, twelve reliability contracts are identified in table 2. The utility offers all of these contracts. Customers select and rank contracts, based upon their own attributes and the characteristics of the contracts. Overall system reliability is determined by the aggregation of the individual choices. The survey data are described in more detail in the appendix.

Page 6: Status quo bias in the measurement of value of service

Tab

le

1

WT

A

and

WT

P m

easu

res

of

valu

e:

Con

tinge

nt

valu

atio

n su

rvey

re

sults

.

Will

ingn

ess

to

acce

pt

(WT

A)

Will

ingn

ess

to

pay

(WT

P)

Scen

ario

T

ime

of

Seas

on

day

Win

ter

Eve

ning

W

inte

r E

veni

ng

Win

ter

Mor

ning

W

inte

r M

orni

ng

Sum

mer

A

fter

noon

Su

mm

er

Aft

erno

on

Sum

mer

A

fter

noon

Any

A

ny

Sum

mer

A

fter

noon

Dur

atio

n N

otic

e

1 ho

ur

Non

e 4

hour

s N

one

4 ho

urs

Non

e 12

ho

urs

Non

e 1

hour

N

one

4 ho

urs

Non

e 12

hou

rs

Non

e M

omen

tary

” N

one

1 ho

ur

Yes

Mea

n M

edia

n “/

,O

Tob

it M

ean

Med

ian

%O

T

obit

“Les

s th

an

five

se

cond

s.

10.7

5 5.

00

24

10.9

6 2.

95

1.00

43

3.

13

19.9

1 10

.00

14

19.5

7 4.

78

2.00

33

4.

16

12.0

8 3.

00

37

9.22

2.

99

0.14

45

3.

01

40.5

5 20

.00

9 33

.15

9.04

5.

00

30

8.99

3.

66

0.00

61

3.

63

1.64

0.

00

63

1.51

13

.59

5.00

36

11

.78

3.60

0.

00

49

3.30

38

.03

20.0

0 9

31.5

6 8.

70

5.00

32

8.

52

1.66

0.

00

69

1.44

0.

16

0.00

69

0.

15

2.79

0.

00

64

3.01

0.

98

0.00

68

1.

01

Page 7: Status quo bias in the measurement of value of service

R.S. Hartman et al., Status quo bias 203

Table 2

Service reliability rate options.

Frequency Average Change in Percent of Option (outages/yr) duration current bill sample choosing”

A: For households with existing reliability characterized by approximately three outages per year

1 3 2 hrs Status quo 60.2% 2 2 1 hrs + 5 percent 13.6% 3 5 2 hrs -5 percent 12.0% 4 5 4 hrs - 10 percent 4.9% 5 15 2 hrs - 20 percent 3.6% 6 15 4 hrs - 30 percent 5.7%

B: For households with existing reliability characterized by approximately 15 outages per year

1 15 4 hrs Status quo 58.3% 2 20 4 hrs - 10 percent 15.1% 3 15 2 hrs + 10 percent 12.7”,’ 4 5 4 hrs + 20 percent 4.7; 5 5 2 hrs + 25 percent 3.4% 6 3 2 hrs + 30 percent 5.8%

“Percent of sample choosing the specific rate option as most preferred.

estimates.8 The ‘corrected’ WTP and WTA valuations corroborate the

sample means. All WTP and WTA valuations are summarized in table 1. As a second test, we use a probabilistic choice model to quantify customer

preferences depicted in tig.l. We estimate the choice model using the data on customer selection of the rate options presented in table 2. These rate options reflect reliability differences, where reliability is defined by the number of power outages of a given duration that occur over a year. Customers therefore have the opportunity to purchase lower reliability at reduced electricity prices (a lower bill) or greater reliability at higher prices

81f the K (individual i’s values of service reliability) are truncated in the sample at 0, then for any random individual i

E(~)=Prob(I/>O)*E(I/;It$>O)+Prob(~=O)*O

=Prob(q>O)*E(I:I K>O).

In the first stage of our Tobit correction, we analyze the probability that an individual’s value of service reliability is positive (Prob( q>O)) using a binary probit specification. In the second stage, we utilize the estimated probit as the truncation-bias correction term (in the form of Inverse Mills ratios) in regressions (E( F 1 F>O)) relating WTP and WTA measures to a

customer’s demographic characteristics, his/her current reliability regime and all attributes of the service. Because this method is fairly standard, we do not develop it here. For details, see Doane, Hartman and Woo (1988b). See also Heckman (1979) Tobin (1958), and Cameron and James (1987), who have recently exploited a similar framework in analyzing truncated data generated by a ‘closed-end’ contingent valuation of consumer willingness to pay for a recreational fishing day. We tested for a second possible truncation from above. The truncation from above was caused when we eliminated outliers in the form of incomprehensibly large cost estimates. This double truncation correction proved unimportant in the WTA and WTP regressions. These results are available from the authors upon request.

Page 8: Status quo bias in the measurement of value of service

204 R.S. Hartman et al., Status quo bias

(an increased bill). In characterizing the rate options, we have carefully described the current level of service reliability as the status quo (point ai).

Our choice-theoretic framework follows standard lines. In particular, we assume that individual i selects his/her most preferred service contract j to maximize random utility, where the utility of contract j to individual i is

Uij= ~fij+ 5ij= BXij+ 5ij, (1)

and Oij is the representative utility obtained by the choice; B is a vector of consumer preferences; Xii is a vector of the attributes of individual i and option j; and tij is a random error term induced by purely random behavior, excluded or improperly measured attributes, or random tastes. Individual i selects reliability option j (j = 1,. . . , J) when Uij> Uik, for all k# j. Hence, the probability that individual i selects j as most preferred is

Prij = prob ( Uij > Uik, all k # j)

= prob (BXij + tij > BX, + 5ik, all k # j)

=prob(5ik-5ij<B{Xij-Xik}, all k# j)

=prob(q,j< B{Xij-Xik}, all k# j), (2)

where u],~= tik- fiJ Assuming the 5ij are distributed normally, ~kj will be normal and the appropriate choice model will be probit. Assuming the tij are distributed as Weibull, logit analysis results.g We make this latter assumption.

We have assumed that customer utility is determined by the reliability of electrical service offered with each service contract, in addition to all other goods and services (measured by income, net of electricity charges). A given reliability contract is characterized by the number of outages expected annually, the expected duration of an average outage and the associated electric bill for each contract [which we denote as frequency (F), duration (D) and cost (C), respectively]. Changes in the electric bill (i.e., changes in income, net of electricity charges) are assumed to summarize demand for other goods and services.

Every customer i is subject to a base service contract (with attributes F,, D, and C,) before being asked to select a preferred alternative. Furthermore,

‘See Hausman and Wise (1978) and Hartman (1982).

Page 9: Status quo bias in the measurement of value of service

R.S. Hartman et al., Status quo bias 20.5

we express representative customer utility relative to the base contract as a function of F, D, C and Zi (the vector Zi summarizes customer attributes):

Ui= i-(F, D, (C/CO), Zi) + do * ALT*, (3)

where the cost effect (C) is measured relative to the status quo monthly bill (C,). We also include AL& as an alternative-specific (or mode-specific) dummy variable denoting the status quo service option (ALTO= 1 when the alternative contract reflects the customer’s current reliability regime; 0 otherwise). This mode-specific dummy allows us to test the hypothesis that there exists a status quo (SQ) effect using a simple t test (H,: SQ =0 if d, =O).

Using (3), representative utility derived from the status quo service reliability is

while the utility for any alternative reliability contract j is

Uij= i7ij(Fj, Dj,(Cj/C*), Zi). (3”)

Given a particular specification of utility equation (3), the definition of reliability options j= 0,. . . , f in table 2, our assumption of Weibull errors tij, and our survey data, we can obtain maximum likelihood estimates of customer preferences. Because we have no strong priors on the shape of utility, we should test a full second-order approximation in F, D, C/C, and Z. Having done so (as reported below), we found that we could not reject the following simplified specification of utility for reliability option j = 0,. . . , J:

Dij=dl*Fj+dZ*Dj+d3* (Cj/Co)+d4*(Cj/CO)*Zi+do*ALTo* (4)

Given estimates of preferences d= {d,, d,, dz, d,, d4), we derive the compen- sation variation required for changes in reliability from the status quo as follows. We seek the com~nsation necessary to make a customer indifferent between the status quo (F,, I),, C,> and the alternative reliability offered by rate option j{Fj,Dj, Cj); in other words, the compensation required for ilij= Ui,.‘O We derive that compensation as follows:

‘“This development is equivalent to the derivation of compensating variation for quality changes in Small and Rosen (1981).

Page 10: Status quo bias in the measurement of value of service

206 R.S. Hartman et al., Status quo bias

dl * Fj + d, * Dj + d, * (Cj/‘C,) + da * (Cj/C,) * Zi

=d, * F,+d, *D,+d,* (Co/C,)+dq*(CO/CO)*Zi+do*ALTo;

(5) (Cj-C,)/C,=(-l)*{d,*(Fj-F,)+d,*(Dj-D,)-d,*ALT,}

lid, +d, * Zi).

(Cj-C,)/C, is the proportional change in the customer’s bill required to compensate him for altered service reliability (Fj- F,) and (Oj-D,). The total compensation required is therefore

TC=(Cj-C())

=(-C,)*{dl *(Fj_F,)+dZ*(Dj_Do)-d,*ALT~}/{d,+d4*Zi}.

(6)

Returning to fig. 1, TC measures the compensating variation required to maintain a customer’s initial level of utility for reliability changes from the status quo, a,, to alternative regimes reflecting decreased reliability, bz, or increased reliability, c2. For increased reliability, the implied compensation will be negative, reflecting a willingness to pay. It should be noticed that given the interactive terms (da), compensation measures will be hetero- geneous in the population.

3. Empirical results

Table 1 indicates that the reported WTA measures are uniformly 3 to 4 times larger than the WTP measures for all outage scenarios.” While the sample truncation evident in the table argues for the two-stage Tobit correction, the ‘corrected’ Tobit measures” are quite similar to the sample means for all outage scenarios except 3 and 4. For all scenarios, the Tobit results corroborate the 3-to-4 time differential between WTA and WTP measures. Both sets of results confirm the kink in household preferences at a, in fig. 1.

Before turning to estimates of our choice model, it is interesting to note in table 2 the percentage of survey respondents in each group that selected the alternative options as their most preferred choice. While there was little difference in the demographic characteristics between the two groups of households, we find that both groups express a strong preference for their

“Some of this difference may reflect a rational distinction. For example, customers may believe that backup service for which they are willing to pay may not be a perfect substitute for regular service.

12The full set of Tobit regressions for WTP and WTA are described fully in Doane, Hartman and Woo (1988b).

Page 11: Status quo bias in the measurement of value of service

R.S. Hartman et al., Status quo bias 207

Table 3

Conditional logit estimation for service reliability choice.“, b

Observations Log-likelihood Log-L (slopes = 0) Chi-squared

FREQUENCY

DURATION

COST

ALT,

COST * AVGOVT

Model 1 Model 2

853 853 - Ll18.3 - 1,104.4 - L528.4 - 1528.4

820.1 845.9

- 0.47306 - 0.32929 (-5.36) ( - 3.49)

- 1.19874 -0.82904 (- 5.97) ( - 3.96)

- 25.92580 -22.41830 ( - 5.97) ( - 4.39)

I .90240 1.71610 (23.48) (19.96)

6.25528 (4.73)

“t-statistics for H,: d = 0 are in parentheses. “Variable Definitions: FREQUENCY = Number of

outages per year. DVRATION=Duration of outage in hours. COST= Bill discount of alternative reliability con- tracts relative to the status quo (equal to the monthly electricity bill of the alternative contract divided by the status quo monthly electricity bill - C,/C,). ALTO= Alternative-specific dummy variable denoting the cus- tomer’s status quo service reliability. ALT, = I when the alternative reliability contract is the status quo; 0 other- wise. AVGOVT=Dummy variable equal to 1 if the household indicated its current reliability level was best characterized by three outages per year; zero otherwise (i.e., household indicated its current reliability level was best characterized by 15 outages per year).

quite different status quo. Our choice model estimates [of eq. (4)] in table 3 reflect these preferences.’ 3 The mode-specific status quo variable (ALT’,) was consistently the most statistically significant determinant of reliability con- tract choice, indicating that residential customers do attach a substantial

r3Based on the chi-squared tests, both models have considerable explanatory power. The coefficients of FREQUENCY, DURATION and COST are highly significant and the correct sign. In both models the effect of DURATION is larger than that of FREQUENCY, implying a single two-hour outage requires a greater compensation than two individual one-hour outages, all else constant. Model 2 introduces the term COST* AVGOVT into Model 1 in order to indicate whether the households’ prior reliability experience affects required compensation. AVGOVT is a dummy variable set equal to one when a household’s current reliability level was best characterized by three outages over the past year (the system average); zero otherwise (i.e., the household indicated its current reliability level was best characterized by 15 outages over the past year). Model 2 indicates that households experiencing a larger number of outages require lower compensation, ceteris paribus. As indicated in the model discussion, we were unable to reject the hypothesis that all other second-order terms were zero.

Page 12: Status quo bias in the measurement of value of service

208 R.S. Hartman et al., Status quo bias

Table 4 Implied compensation for altered reliability levels.’

Attributes TC by monthly bill quantile From base --- - to option Freq. Dur. 10% 25% 50% (median) 75% 90%

A: including mode-spec~~c ‘status quo’ eflect, % mo~rh

2 2 1 0.31 0.83 1.60 3 5 2 1.57 3.54 6.81 4 S 4 2.68 6.02 11.57 5 15 2 3.76 8.46 16.25 6 1s 4 4.86 10.94 21.01

B: Netting out the mode-spec$c ‘status quo’ e&cr, $ month

2 2 1 -0.76 - 1.73 - 3.32 3 s 2 0.44 0.98 1.89 4 s 4 1.53 3.46 6.64 5 1s 2 2.62 5.90 11.34 6 15 4 3.72 8.38 16.09

2.24 3.13 9.56 13.36

16.24 22.69 22.82 31.89 29.50 41.22

-4.66 -6.52 2.65 3.71 9.33 13.04

15.91 22.24 22.59 31.57

“Base option attributes: Freq.= 3; Dur.=2. Freq.=number of outages per year. Dur.=average duration of outage in hours. Average monthly electricity bill quantiles: lO%=rSil.lS; 25%=$25.08; 50%=%48.16; 75%=$67.61; 90%=$94.48. All values are in 3986 dollars. Negative compeasat~on estimates imply willingness to pay.

premium to their existing service level, ceteris paribus, and that they have a strong aversion toward alternative reliability options no matter how desir- able they may be based upon attributes (F and D) and cost (C).14

Using the parameter estimates from choice model 2, table 4.A reports the compensating variation required for a move from a status quo of three outages of two hours each per year (option 1) to the alternative reliability options presented in table 2.A. Because our estimates are conditional upon the size of a households status quo monthly electricity bill (C,), we present compensation estimates (TC) for five monthly bill quantiles (lo%, 25%, 50’+& 75x, 90%) where the 50% quantile is defined as the median household.

The compensation estimates in table 4.A are derived using eq. (6); they include the mode-specific status quo effect. The compensation estimates in table 4.B net out the mode-specific status quo variable by setting d,=O in eq. (6). A comparison of these two sets of estimates is informative. Netting out the mode-specific ‘status quo’ effect, the median household (50% quantile) in table 4.B would be willing to pay $3.32/month15 to move from its current reliability level (three outages of two-hour duration, annually) to an i~c~eus~~

L4Alternative-specifc dummy variables for the other options proved insignificant. This predisposition for the status quo may result from familiarity and satisfaction with the current level of service; a belief that the utility will not be able to provide the actual level of service offered by the rate option; habit and/or inertia.

15We interpret a negative ~m~nsation as a winningness to pay.

Page 13: Status quo bias in the measurement of value of service

R.S. Hartman et al., Status quo bias 209

Table 5

A comparison of self-stated WTA and required compensation (TC) for identical decreases in reliability ($/outage).”

Self-stated Total compensation including cost (WTA) mode-specific effect

One-hour outage 7.29 52.78

“The one-hour outage WTA represents the average costs of scenarios I and 5 in table 2. The total compensation eq. (6) is calculated for the median household using model I in table 3. The calculation assumes that the household’s status quo reliability of three outages of one-hour duration per year was reduced to four outages of one-hour duration per year. All values are in 1986 dollars.

reliability level (two outages of one hour, annually); this same household would require compensation of $1.89 or $16.09/month to move from its current reliability level to the diminished reliability of options 3 (five outages of two hours, annually) or 6 (15 outages of four hours, annually), respectively.

When we include the mode-specific status quo effect (table 4.A), however, this same median household would require a compensation of $1.60/month to move to the improved reliability level offered in option 2, precisely because the disutility of leaving the status quo outweighs the perceived benefit of the improved service reliability. Similarly, this same household would require compensation of $6.81 or $2l.Ol/month to move to the diminished reliability levels of option 3 or option 6, respectively. For all scenarios in table 4, the compensation required for rate switching from the status quo is found to consistently be much higher than those levels required if there were no mode-specific status quo inertia. In fact, the customers must be compensated for switching reliability regimes even when the alternative regime entails more reliable service.16 This suggests that the kink in utility at a, in fig. 1 is even more serious than that suggested by the disparity between WTA and WTP. Because customers must be compensated {or small increases in reliability from the status quo, utility I’ rather than I, is a better representation of preferences for reliability increases immediately to the right of a,.

Furthermore, the required compensation for reliability decreases to the left of a,, as suggested by the choice analysis, are higher than those suggested by the self-stated WTA. Evidence supporting this contention is provided in table 5 for a scenario of diminished reliability: a single one-hour outage. The sample mean WTA for the one-hour outage is $7.29 while the compensation implied by the choice model (including the mode-specific status quo effect) is $52.78. These estimates suggest that I’ may also be a better representation of

16Parenthetically, table 4 also indicates a wide dispersion of implied compensation, a result corroborated in the Tobit analysis. For example, households in the 90% monthly bill percentile require compensation approximately ten times larger than those households in the lOu/, percentile. This suggests considerable customer heterogeneity.

Page 14: Status quo bias in the measurement of value of service

210 R.S. Hartman et al., Status quo bias

consumer utility for rehability decreases to the left of a, in fig. 1. Clearly, I’ is more kinked at o1 than 1,.

4. Summary and conclusions

Our analysis indicates that status quo effects can introduce serious distortions into the welfare measures routinely exploited to assess the social desirability and customer acceptance of priority service. We find large disparities (3 to 4 times) between the WTA and WTP measures for all outage scenarios, con~rming the existence of the hypothesized kink (along I,) in utility at the status quo. Our analysis of the choice of reliability regime suggests that the kink is even more severe (along I’). As a result, compensa- tion levels required for reliability decreases are found to be considerably higher than WTA estimates. Furthermore, customers are not willing to pay for marginal reliability increases; rather, they require compensative for reliability increases from the status quo.

Such vastly disparate WTA, WTP and compensating variation measures of value offer quite equivocal policy signals. We conclude that, in order to formulate and implement policy regarding priority service, detailed infor- mation regarding ail of these valuations is necessary. Development of the complete set of valuations must be viewed as the first step for utilities interested in designing new products and services or considering adding capacity to improve the existing service quality. Once the complete set of valuations is developed, they can be compared, and different valuations can be used for different analyses.

For example, the WTP measures are the most conservative and seem to offer a reasonable lower bound for traditional welfare calculations. This interpretation is supported by experimental information, such as the work of Coursey, Hovis and Schulze (1987). These authors argue that the WTA and WTP measures developed in many contingent valuation surveys are elicited from respondents for hypothetical, potentially-unfamiliar commodities. They believe that observed three-to-one disparities between WTA and WTP measures reflect the essential novelty and one-time nature of the choices rather than actuai preferences. To test this hypothesis, they develop a repeated bidding experiment that allows for the respondents to learn about an unfamiliar good. They find that initial WTA valuations are orders of magnitude (three-to-one) greater than WTP measures. However, they also find that the respondents’ WTA measures decline and converge to their WTP measures in the series of repeated bidding experiments. Such results suggest that WTA and WTP measures may be approximately equal when valuing

familiar goods and services and that WTP measures are preferable overall.”

“Brookshire and Coursey (1987) corroborate these results.

Page 15: Status quo bias in the measurement of value of service

RX Hartman et al., Status quo bias 211

Our analysis of customer choice of reliability rate option sheds limited additional light on this learning hypothesis. The results in table 3 suggest that customer experience with outages lowers the compensation required for diminished reliability. If more rational valuations can be obtained through learning, it may be appropriate to ask electricity customers to select their reliability contract every year.

Our analysis of customer choice furthermore indicates that knowledge of WTA and WTP measures may still prove insufficient for predicting market penetration of the newly-designed rate options. Such estimates may seriously over-estimate consumers’ willingness to accept alternative reliability- differentiated service options because of the status quo effect.18 Explicit analysis of the consumer choice process is required to measure and design the incentives necessary to overcome inertial status quo effects.

Appendix: The data

Our data were obtained through a survey questionnaire mailed to a stratified random sample of 2200 residential customers in the Pacific Gas and Electric Company (PG&E) service area. Table A.1 presents the sample means of customer and demographic characteristics for the approximately 1500 survey responses. The survey was stratified by geographical location to allow adequate urban/rural and climate zone variation. A high response rate (approximately 70%), a subsequent non-respondent telephone survey, and a comparison of the customer demographic data with other PG&E surveys indicate the survey data are representative. [See Doane, Hartman and Woo (1988a, b)].

The survey described the nine outage scenarios presented in table 1, which also summarizes the mean, median and truncation at zero. The mean is computed for customer-reported costs below the 99.5 percentile, in order to remove atypical highly influential customer responses (see footnote 7) which seem to be outliers (e.g., households which reported identical cost estimates for each of the different scenarios while not completing the rest of the survey).

The two menus of alternative reliability options offered to respondents is presented in table 2. The first menu was offered to those customers who had experienced approximately three outages of two-hours duration during the last year (90% of the sample). The second menu was offered to households that had experienced approximately 15 outages of four-hours duration during the preceding year. All respondents were asked to rank the options

“This phenomenon is not new. The need to offer low ‘introductory’ prices for new products to overcome status quo inertia is a recognized business strategy. Ignoring status quo inertia can be serious. The greatest marketing error in recent decades - the substitution of ‘new’ for ‘old Coca Cola - stemmed from a failure to recognize status quo bias. See ‘Saying No to New Coke’, Newsweek, June 24, 1985, pp. 32-33.

Page 16: Status quo bias in the measurement of value of service

212 R.S. Hartman et al., Status quo bias

Table A.1

Sample mean of customer characteristics.

Variable N” Mean - Income Household size (persons) Average age (years) Number of household members

65 years or older

Average monthly electricity sales in winter (kwh)

Average monthly electricity sales in summer (kwh)

Rural Bay area Large city

Electric applicance ownership Space heater Water heater Central air conditioner Window air conditioner Range Security alarm Personal computer Video cassette recorder

Home business

Health problem

Number of household members generally at home during the day

Number of outages experienced during last I2 months

aN = Number of observations.

1,281 $33,648 1,488 2.71 1,488 41.0

1,488 0.37

1,501 511

1.501 495

1,460 0.13 1,509 0.71 1,460 0.32

1.435 0.20 1,470 0.17 1,498 0.23 1,432 0.15 1,474 0.64 1,427 0.12 1,433 0.16 1,430 0.55

1,451 0.10

1,449 0.07

1,445 1.29

1,408 2.90

Standard deviation

$20,177 1.43

17.1

0.68

371

377

0.34 0.45 0.47

0.40 0.38 0.42 0.35 0.48 0.32 0.37 0.50

0.30

0.26

1.15

4.33

from least to most preferred. The options were configured around the status quo reliability to represent realistic service alternatives. The questionnaire was carefully designed to avoid strategic response bias. In particular, respondents were reminded that the reliability provided by PG&E helps to determine the cost of electrical service. They were told that while PG&E cannot prevent all causes of power outages, it could spend more money to improve service, which could increase rates, or it could reduce reliability and possibly reduce rates. Therefore, information on their reliability preferences for different levels of service would be used to help plan future service. The rate options presented in table 2 were listed in different orders to different respondents.

References

Bishop, R.C. and T.A. Heberlein, 1979, Measuring values of extra-market goods: Are indirect measures biased?, American Journal of Agricultural Economics 7, 916930.

Page 17: Status quo bias in the measurement of value of service

R.S. Hartman et al., Status quo bias 213

Bohn, R.E., M.C. Caramanis and F.C. Schweppe, 1984, Optimal pricing in electrical networks over space and time, The Rand Journal of Economics 15, no. 3.

Brookshire, D.S. and D.L. Coursey, 1987, Measuring the value of a public good: An empirical comparison of elicitation procedures, American Economic Review 77, no. 4, 554566.

Cameron, T.A. and M.D. James, 1987, Efficient estimation methods for ‘closed-ended’ contingent valuation surveys, Review of Economics and Statistics 69, no. 2, 269-276.

Chao, H.P. and R. Wilson, 1987, Priority service: Pricing, investment and market organization, American Economic Review 77, no. 5, 899-916.

Coursey, D.L., J.L. Hovis and W.D. Schulze, 1987, The disparity between willingness to accept and willingness to pay measures of value, Quarterly Journal of Economics 102, no. 3, 679-690.

Cummings, R.G., D.S. Brookshire and W.D. Schulze, eds., 1986, Valuing environmental goods: An assessment of the contingent valuation method (Rowan and Allanheld, Totowa, NJ).

Doane, M.J., R.S. Hartman and C.K. Woo, 1988a, Household preference for interruptible rate options and the revealed value of service reliability, Energy Journal - Special Issue.

Doane, M.J., R.S. Hartman and C.K. Woo, 1988b, An econometric analysis of perceived value of service reliability, Energy Journal - Special Issue.

Freeman, A.M., III, 1982, Air and water pollution control: A benefit-cost assessment (Wiley, New York).

Friedman, M. and L. Savage, 1948, The utility analysis of choices involving risk, Journal of Political Economy 56, 279-304.

Hartman, R.S., 1982, A note on the use of aggregate data in individual choice models: Discrete consumer choice among alternative fuels for residential appliances, Journal of Econometrics 18, 313-335.

Hartman, R.S. and M.J. Doane, 1986, Household discount rates revisited, The Energy Journal 7, no. 1, 139-148.

Hartman, R.S. and C.K. Woo, 1988, The value of service reliability: Alternative welfare measures, Working paper no. 167 (Department of Economics, Boston University, Boston, MA).

Hartman, R.S., M.J. Doane and C.K. Woo, 1990, Consumer rationality and the status quo, The Quarterly Journal of Economics, forthcoming.

Hausman, J.A., 1979, Individual discount rates and the purchase and utilization of energy-using durables, Bell Journal of Economics 10, no. 1, 33-54.

Hausman, J.A. and D. Wise, 1978, A conditional Probit model for qualitative choice: Discrete decisions recognizing independence and heterogeneous preferences, Econometrica 46, 403426.

Heckman, J., 1979, Sample selection bias as a specification error, Econometrica 47, 1533161. Kahneman, D. and A. Tversky, 1979, Prospect theory: An analysis of decision under risk,

Econometrica 47, no. 2, 263-291. Knetsch, J.L. and J. A. Sinden, 1984, Willingness to pay and compensation demanded:

Experimental evidence of an unexpected disparity in measures of value, Quarterly Journal of Economics 99, no. 3, 507-521.

Knez, P., V.L. Smith and A. Williams, 1985, Individual rationality, market rationality and value estimation, American Economic Review 75, 397402.

Machina, M., 1987, Choice under uncertainty: Problems solved and unsolved, Economic Perspectives 1, no. 1, 121-154.

Mitchell, R.C. and R.T. Carson, 1981, An experiment in determining willingness to pay for national water quality improvements, draft report prepared for the U.S. Environmental Protection Agency (Resources for the Future, Inc., Washington, DC).

Munasinghe, M., 1979, The economics of power system reliability (Johns Hopkins University Press, Baltimore, MD).

Munasinghe, M. and M. Gellerson, 1979, Economic criteria for optimizing power system reliability levels, Bell Journal of Economics 10, no. 1, 353-365.

Oren, S., S. Smith, R. Wilson and H. Chao, 1986, Priority service: Unbundling the quality attributes of electric power, EPRI Report EA-4851 (Electric Power Research Institute, Palo Alto, CA).

Page 18: Status quo bias in the measurement of value of service

214 R.S. Hartman et al., Status quo bias

Rowe, R.D., R.C. d’Arge and D.S. Brookshire, 1980, An experiment on the economic value of visibility, Journal of Environmental Economics and Management 7, no. 1, l-19.

Samuelson, W. and R. Zeckhauser, 1986, Status quo bias in individual decision making, Working paper (Harvard University, Cambridge, MA).

Scherr, B.A. and E.M. Babb, 1975, Pricing public goods: An experiment with two proposed pricing systems, Public Choice 23, 35-48.

Schulze, W.D., R.C. d’Arge and D. Brookshire, 1981, Valuing environmental commodities: Some recent experiments, Land Economics 57, no. 2, 151-172.

Small, K.A. and H.S., Rosen, 1981, Applied welfare economics with discrete choice models, Econometrica 49, no. 1, 105-130.

Smith, V.K. and W.H. Desvousges, 1987, An empirical analysis of the economic value of risk changes, Journal of Political Economy 95, no. 1, 89-114.

Smith, V.L., 1979, Incentive compatible experimental processes for the provision of public goods, in: V.L. Smith, ed., Research in experimental economics, Vol. 1 (JAI Press, Greenwich, CT).

Tobin, R.P., 1958, Estimation of relationships for limited dependent variables, Econometrica 26, 2436.

Tschirhart, J. and F. Jen, 1979, Behavior of a monopoly offering interruptible service, The Bell Journal of Economics 10, no. 1.

Willig, R.D., 1976, Consumers’ surplus without apology, American Economic Review 66, 589-597.