Economics Alumni Working Paper Series
Broker Duty to Clients: Why States Mandate Minimum Service
Requirements
Anupam Nanda
Mumbai
Katherine A. Pancak
University of Connecticut
Working Paper 2009-01
April 2009
341 Mansfield Road, Unit 1063
Storrs, CT 06269–1063
Phone: (860) 486–3022
Fax: (860) 486–4463
http://www.econ.uconn.edu/
This working paper is indexed on RePEc, http://repec.org/
Abstract
Since 2004, ten U.S. states have enacted laws that mandate real estate brokers
to provide real estate consumers with a minimum level of services. The federal
government and the academic literature suggest that such state laws are a result
of anti-competitive industry collusion, and serve no consumer protection justification. This paper attempts to determine the factors that led states to adopt minimum service requirements, despite significant federal opposition. The analytical
structure employs hazard models, using a unique set of economic and institutional
attributes for 50 U.S. states from 2000 to 2007. Contrary to initial expectations
based on the literature, our results indicate that both strength of a state.s Realtor association and broker membership on real estate licensing boards decrease,
rather than increase, the likelihood of state adoption of broker minimum service
requirements. Factors that do increase the likelihood of adoption include higher
state licensing complaints and a democratically controlled state legislature.
Journal of Economic Literature Classification: C14, K11, L85, R21
Keywords: Minimum Services, Hazard Model
Authors acknowledge helpful comments from John Clapp. All remaining errors are ours.
1. Introduction
A number of states have recently enacted laws requiring a real estate broker i to provide a
real estate consumer (buyer, seller, landlord, or tenant) with a minimum level of services,
including requirements to assist in negotiating, present and receive offers, and answer questions.
The United States Department of Justice Antitrust Division (DOJ) and the United State Federal
Trade Commission (FTC) oppose these types of requirements on the grounds that they are anticompetitive, and have lobbied heavily against state enactment. These types of laws are deemed
anti-competitive primarily because they prevent a limited service real estate broker from
contracting with a seller to only provide access to the MLS for a flat fee.
The purpose of this paper is to try to determine what factors led states to enact minimum
service laws, despite significant federal opposition. The analytical structure employs hazard
models, using a unique and rich set of economic and institutional attributes for the housing
market in a yearly panel of 50 US States from 2000 to 2007. Among the factors analyzed, it
appears that strength of state association and broker presence on state licensing boards decrease
the likelihood that state minimum service laws are adopted. State level of complaints against
licensees, legislature partisanship, prelicensing requirements, and higher level of population
growth on income increase the likelihood of adoption.
Organization of our paper is as follows. The next section discusses prior literature on
minimum service laws. The third section reviews the various categories of minimum service
laws, and looks at legislative background of states that have passed laws that require brokers to
provide consumers with a minimum level of services. The fourth section presents the institutional
and economic variables used. Here we also make some hypotheses of the relationships between
minimum service law adoption and the independent variables we have chosen. The fifth section
presents the empirical method that we employ in this study. The sixth section analyzes, compares,
2
and contrasts the results from different model specifications. In the final section, we summarize
our findings and suggest potential state legislative policy trends.
2. Prior Literature
The DOJ and FTC compiled a report entitled “Competition in the Real Estate Brokerage
Industry”, published in April 2007 (FTC and DOJ 2007). One part of the report discusses the
effect of minimum service requirements on brokerage competition, defining the minimum service
requirements as “laws and regulations that enumerate specific tasks that a broker must perform
for a client.”ii Missouri brokerage law is given as an example. In that state, all brokers entering
into an exclusive brokerage agreement must (1) accept delivery of and present offers and
counteroffers to clients and customers; (2) assist clients and customers in developing,
communicating, negotiating, and presenting offers, counteroffers, and disclosure notices; and (3)
answer clients and customers questions relating to offers, counteroffers, notices, and
contingencies.
The DOJ and FTC strongly assert that such minimum service requirements are anticompetitive because they reduce choice of real estate brokerage services, force real estate
consumers to buy services they may not want, and block limited-service brokers from offering
less than a full package of real estate brokerage services. While the report review the claims of
supporters that such laws protect both consumers and brokers, the DOJ and FTC conclude that
there is no evidence to date that consumers benefit from minimum service laws.
There is very little academic real estate literature on real estate broker minimum service
laws. A handful of authors have looked at the issue tangentially when analyzing whether real
estate brokerage is competitive. Hahn, Litan, and Gurman (2006) identified state legislation
establishing minimum service requirements as one impediment to competition. Based on a review
3
of the academic literature and interviews with real estate industry participants, the U.S.
Government Accountability Office concluded that widespread use of the internet in real estate
transactions should encourage more brokerage price variation, but may be hindered by obstacles
including minimum service laws (GAO 2006). Miceli, Pancak, and Sirmans (2007) determined
that minimum service laws may be an attempt to prop up a brokerage compensation scheme that,
while possibly in the best interests of brokers, is socially unproductive. Magura (2007) proposed
that state minimum service laws have a chilling effect on broker price cutting by accommodating
broker steering behavior. White (2006) observed that mandatory minimum service requirements
for sellers’ brokers eliminated competition from discount brokers whose only service would be to
provide access to a MLS.
Levitt and Syverson (2008) analyzed whether there are any consumer protection
justifications for broker minimum service laws. They compared variables for houses listed with
limited service brokers and those with full service brokers. They found that houses listed with
limited service brokers take longer to sell but eventually sell for at similar prices to those listed
with full-service brokers. They weighed the trade-off between the lower fees charged by a
limited-service broker and the longer time on the market, and reasoned that consumers using
limited-service flat-fee brokers were not worse off than those using full-service, full-commission
brokers. Based on this analysis, they concluded that brokerage minimum service laws are not
needed to protect consumers.
Pancak (2008) looked at the substance of specific state brokerage laws that could be
interpreted as requiring a minimum level of services. She identified a wide range of regulatory
provisions in the fifty states, and compared all state provisions to the DOJ’s website listing states
with minimum service laws.iii She found that the DOJ list was incomplete and incorrect. She
points out that one of the primary weaknesses with the DOJ list is that some states with language
4
requiring brokers to “present offers in a timely manner” are included as part of the anticompetitive list, while others are not. Without clear legislative history to the contrary, she
concluded that these types of provisions should be interpreted as requiring timely communication
when an offer is transmitted to the broker, not requiring that brokers be available to accept and
present offers.
Pancak determined the actual level of brokerage minimum service laws in all states,
categorizing them into four categories: states with minimum service requirements in all brokerage
relationships, states with minimum service requirements in certain brokerage relationships, states
requiring timely communication of offers, and states with waivable minimum service
requirements and/or requiring timely communication of offers. The data presented in the Pancak
article is used as the basis for the dependent variable in this paper.
3. Enactment of State Minimum Service Laws
Evidence presented below indicates that the primary supporter of the state minimum
services law has been state Realtor associations. The National Association of Realtors (NAR) has
not formally endorsed state minimum-service laws, noting that some state associations are in
favor of such laws while others are not.iv NAR has, however, offered legal advice about such
legislation, noting that associations have the right to propose and lobby for laws that they support,
even if the law may be deemed by federal agencies to be anti-competitive.v
Minimum service law supporters maintain that real estate consumers need to be able to
expect certain services from a broker hired to represent them, and laws are needed to guarantee
that minimum level of services (DOJ and FTC 2007). Additionally, if a seller working with a
limited service broker does not receive assistance from that broker, the seller may need to ask
assistance from the buyer’s broker. They also reason that the buyer’s broker needs to be protected
5
from having the additional work of assisting a seller, since the buyer’s broker is not paid by the
seller and helping the seller may create a dual agency conflict. Finally, some are concerned that a
buyer broker may not be compensated by a seller working with a limited-service broker who
decides to deal directly with a buyer.
The DOJ and FTC have taken an aggressive stance against minimum service laws and
proposed laws. The DOJ website lists its efforts to eliminate minimum service laws, including
links to numerous press releases and letters to state governors, legislatures, and real estate
commissions.vi The DOJ and FTC can only urge states not to pass laws or to change current laws,
since state legislatures (and boards acting according to legislative intent) may not be sued by the
federal government for federal antitrust violations.vii The state agencies efforts have been met
with mixed reactions. Alabama, Idaho, Missouri, and Texas passed minimum service laws despite
active federal opposition. New Mexico, Tennessee, and Michigan changed proposed legislation to
make minimum service waivable after lobbying by the federal agencies.
Illinois was the first state to adopt minimum service requirements in 2004. Now, all
exclusive brokerage arrangements entered into between a broker and client must specify that the
broker will provide the following services:
1. accept delivery of and present to the client offers and counteroffers to buy, sell, or lease
the client’s property or the property the client seeks to purchase or lease;
2. assist the client in developing, communicating, negotiating, and presenting offers,
counteroffers, and notices that relate to the offers and counteroffers until a lease or
purchase agreement is signed and all contingencies are satisfied or waived; and
3. answer the client’s questions relating to the offers, counteroffers, notices, and
contingencies.
6
There was no specific discussion of the need for this new provision during the state Senate
consideration.viii The Illinois Association of Realtors reported the purpose of the requirement is to
“promote greater professionalism and accountability within the industry and to provide greater
protections for real estate consumers.”ix
An article in Chicago Agent, entitled “The Value of Full Service”, provides additional
insight into Illinois real estate professionals’ opinion that minimum service requirements are
needed.x The article states that the most common complaint from full-service brokers who have
done deals with limited service brokers is that there is no other professional representing the other
side so that the broker “feels the burden of doing both sides, and that doesn’t feel fair or right.”
The brokers feel that this predicament is unfair financially, since they have to do more work in
this situation for the same commission.
Other state’s adopting minimum service requirements echo these types of consumer
protection and broker protection rationale. The Texas Association of Realtors supported adoption
of the minimum-services law in Texas stating that “it would help ease confusion in real estate
transactions and provide true consumer protections." In testimony before the Texas Real Estate
Commission, the Chairman of the Association spoke in favor of the rule, noting that he was
representing 70,000 members. He cited reports of sellers left confused and not helped by the
sellers’ limited service brokers, and of buyer brokers having to step in and help.xi
The Alabama Real Estate Commission stated that the intent of its 2005 law was to limit
MLS-listing-only brokerage activity so that sellers would not be left on their own in transactions
and have no one to answer their questions.xii Iowa law specifying content of brokerage
agreements was passed by the Iowa Legislature in 2005.xiii There are very little legislative
7
comments about why the bill passed; Iowa Real Estate Commission minutes only statement about
the bill was that it was modeled after Illinois.xiv
In addition to an administrative regulation that already required a broker to use
reasonable care in conveying information to a client, Arizona added language in 2005 imposing
an affirmative obligation to take reasonable steps to assist a client in confirming the accuracy of
the information. While not explaining why they made this change, the Arizona Department of
Real Estate stated that brokers must now actively advise clients, and not just tell clients “I don’t
know, you figure it out.”xv
An article reporting on the 2005 Missouri minimum services laws, credits it passage as
being due to heavy lobbying efforts by the Missouri Association of Realtors.xvi Purportedly, the
Association hired a lobbyist for $50,000 to persuade the Missouri governor not to sign the bill
that passed both the Missouri House and Senate unanimously, despite requests from the DOJ and
FTC to veto it.
Kentucky passed an administrative regulation in 2005 that requires brokers representing
a client to perform specific services, including assisting clients in developing, communicating,
negotiating, and presenting offers, and answering questions relating to offers. A broker’s failure
to comply with the minimum requirements is considered gross negligence. It’s interesting that
this regulation is still on the books in Kentucky, as the Kentucky state legislature contemplated
enacting a state statute requiring similar services in 2006. After a letter from the DOJ to the state
legislature, the minimum service language was taken out of the bill.xvii Regardless of the fact that
the legislature did not pass minimum services into a state statute, the administrative regulation
still makes it a provision that brokers have to abide by, and therefore preclude a consumer from
buying MLS-listing-only service.
8
Idaho enacted a law in 2007 that requires any broker entering into a written
representation contract must “be available” to the client to receive and timely present offers. On
its face, this language alone does not necessarily appear to require minimum services. In many
states, the general understanding of the intent of this type of language is to mandate prompt
communication rather than require service be performed. For example, the Oklahoma Attorney
General found that language requiring a broker "being available" to receive all offers does
preclude a broker and client from agreeing that the client receive offers directly.xviii The Idaho
Real Estate Commission, however, interprets the Idaho law as mandating that brokers receive and
review all offers, even though a broker doesn’t have to provide advice to clients or negotiate on
their behalf.xix The bill proposing this law states that its purpose is to clarify that receiving and
presenting offers is a duty that brokers owe clients. While the Idaho Real Estate Commission was
considering its own draft legislation, the proposal that passed was sponsored by the Idaho
Association of Realtors, and the contact name on the bill is the Association’s Director of
Government Affairs.xx
4. Data
Since 2004, ten states have enacted laws that require brokers to offer some minimum
level of service in either all or in exclusive broker relationships. These states, along with
enactment dates and relevant statutory or administrative regulatory provisions are listed in Exhibit
1. We looked at both institutional and economic variables that may have had an effect on the
likelihood of adoption for all fifty states from 2000 to 2007 (408 observations). The time period
allows us to observe most of the states sufficiently before and after the adoption of the minimum
services requirement. Exhibit 2 presents the variables used in our analysis. Exhibit 3 reports the
summary statistics of variables used in this study.
9
As reported by Pancak (2008), four other states also have laws requiring brokers to
provide clients with some type of minimum level of services, but these laws were enacted prior to
2000.xxi These states are also listed in Exhibit 1. Given that the anti-competitive concern about
precluding limited service brokers has only developed in the past decade, we assume that laws
enacted prior to 2000 were driven by factors different than those driving the laws enacted since
2004.xxii As a check of robustness, we also ran the analysis taking out those four states from the
sample. As suspected, the change in magnitude, sign, and standard errors is negligible.
What factors led ten states to revise their licensing laws to require that brokers provide
real estate consumers with a mandatory minimum level of services? We formulate and test
various hypotheses regarding institutional and economic factors in housing market that influence
law enactment of minimum services requirement. Institutional factors related to real estate
licensing are of particular interest.
4.1 Institutional Variables
Most state institutional variable statistics used are derived from the Digest of Real Estate
Licensing Laws and Current Issues (reports from 1999 till 2007) compiled by the Association of
Real Estate Licensing Law Officials (ARELLO). Statistics from other sources are noted in
Exhibit 2.
The purpose of real estate licensing laws is to protect real estate consumers. For that
reason, a state experiencing a high level of real estate consumer complaints relative to number of
state licensees may look to laws to clarify real estate agency relationships and duties. Therefore,
we expect the volume of complaints filed with a state real estate licensing board to directly
influence the likelihood of a state enacting minimum service laws. We control for state size by
looking at complaints per 100 licensees in our analysis.
10
The DOJ and FTC have accused states of enacting minimum service laws at the urging of
state Realtor associations, solely to protect the traditional brokerage fee structure.xxiii We therefore
test the strength of a state’s Realtor association effect on the likelihood of a state passing a
minimum service law. If the federal agencies are correct, we would expect that an increase in the
percentage of state NAR membership equates to an increase in the likelihood that a state would
enact minimum service laws. However, just as possible, a strong associate may actually want to
encourage broker competition and may oppose minimum service requirements on anticompetitive grounds.
The Consumer Federation of America has speculated that there is a connection between
minimum service laws and the number of brokers serving on state real estate boards.xxiv To test
this assertion, we look at the effect of the percentage of brokers or salespersons serving on a state
board, expecting that a higher percentage increases the likelihood that the board will enact
minimum service regulations or encourage state legislatures to pass minimum service statutes.
Ease of entry into the real estate brokerage profession in a state may be a concern for
traditional brokers that fear competition from new limited service brokers. A state’s prelicensing
education hours is an indication of licensing ease or difficulty. Also, a state’s continuing
education hours can be an indication of cost to maintain a license. We expect a state’s
prelicensing hour requirement and continuing education hour requirement to be inversely related
to the likelihood that a state enacts minimum service laws: the easier it is to enter or stay in the
profession, the more likely full service brokers may see limited service brokers as a threat.
Many states have a consumer protection fund which serves to compensate a consumer
that is unable to collect a monetary judgment against a licensee. The dollar amount limit varies by
state. We expect a higher dollar limit to encourage states to enact minimum service laws given
11
that boards and legislatures may fear depleting funds without these perceived additional consumer
protection measures.
The political climate of a state often influences new legislation. We therefore include a
variable for partisan control in the state legislation. Based on common political perceptions, a
state that is controlled by a democratic legislature may be more likely to pass greater consumer
protection legislation.
4.2 Economic Variables
We also control for economic variables in our analysis by including some fundamental
measures that influence the housing market. To measure the housing price changes, we use the
repeat sales purchase-only quarterly Housing Price Index (HPI), reported by the Office of Federal
Housing Enterprise Oversight (OFHEO). We take the average year-over-year rate of change for
the year. Among the fundamental drivers of the housing market, we use percent change in
housing transactions, population growth rate, percentage change in per capita income, and
percentage change in Gross State Product (GSP). The data is obtained from the National
Association of Realtors (NAR), US Census Bureau, US Bureau of Labor Statistics, and US
Bureau of Economic Analysis. Weak state broker productivity may cause traditional full service
real estate brokers to feel the need to protect their business activity from encroachment by limited
service brokers. If this is the case, then we expect that variables potentially causing lower broker
productivity would increase the likelihood of state enactment, and variables resulting in high
broker productivity would decrease the likelihood. We use the number of state residential
transactions per year per real estate licensee as an indication of state broker productivity.
Likewise, a decrease in any one of the other economic variables may threaten real estate
broker activity, encouraging a protectionist position that supports reducing brokerage competition
12
by restricting limited service brokers. We thus expect to see an inverse relationship between the
change in any one of these variables and the likelihood of a state adopting minimum service laws.
5. Empirical Model
Taking into account both the institutional and economic variables set forth in the
preceding section, and following Kiefer (1988), de Figueiredo and Vanden Bergh (2004), and
Nanda (2008), we employed a discrete time proportional hazard model to ascertain which factors
led states to enact mandatory broker minimum service laws. We model the law adoption process
by specifying a probability distribution for the survival spell until “death”, which here is law
enactment.xxv
The probability distribution is given by
F (t )  Pr(T  t )
(1)
which specifies the probability that the random variable T is less than some value t; f(t) is the
corresponding density function. The hazard function can be represented as
 0 (t ) 
f (t )
d ln S (t )

S (t )
dt
(2)
 0 (t ) is the rate at which spells will be completed at duration t; given that they survive until t.
A proportional hazard framework with time-invariant regressors can be written as
 [t ; x]  k[ x] 0 (t )
(3)
where k ( x)  0 is a nonnegative function of x ; and  0 (t )  0 is the baseline hazard.
Time is separated from the explanatory variables so that the hazard is obtained by shifting the
baseline hazard (which is common to all units) as the individual hazard function changes based
13
on a function k (x) of observed covariates (i.e. for all the cross section units, the hazard is
proportional to the baseline hazard function). k ( x)  0 is parameterized as follows:
 [t ; x]  exp[ x ] 0 (t )
(4)
The baseline hazard can be specified by including a function of time. Since no state has ever
repealed the minimum service requirements, we censor the data to the pre-adoption levels. The
hazard model in equation (4) is estimated after conditioning on the event not yet having occurred
using a standard logit specification. We can also incorporate time-varying covariates into the
framework to obtain a conditional hazard function as follows:
 [t ; x m , ]  k[ x m ,  ] m ,
m  1,........, M
(5)
where  is a vector of unknown parameters.
Equation (5) implies that time-varying covariates have a multiplicative effect in each time
interval (for M intervals) and it allows a flexible baseline hazard, which is common to all units.
Incorporating time-varying covariates is shown to have significant impact on the interpretation of
results as the law has been adopted at different times by different states (see Nanda (2008)). Since
different states are likely to have different distributions of the duration dependence and some
relevant and important unobservables (i.e. a potential omitted variable bias) may exist, we control
for the state-level heterogeneity. As shown in Nanda (2008), heterogeneity can be incorporated
into our framework as follows:
 [t ; v, x m , ]  vk[ x m ,  ] m ,
a m 1  t  a m
(6)
where v  0 is a continuously distributed heterogeneity term.
We try the most common distributions for specifying heterogeneity (or frailty), which is the
gamma distribution.
14
6. Results
We looked at enactment of minimum service laws with both time-invariant and timevarying variables. Typically, the pre-treatment averages or time-invariant variables are used in
hazard models with a constant or flexible baseline hazard. However, since minimum service
requirements were enacted in different states at different times, i.e. a disparate treatment due to
state level heterogeneities exists, we try to bring in more information to our estimation system by
incorporating time-varying attributes.
Although pseudo-R2 is not to be taken as having R2-like connotations, our logit and probit
models (with pseudo-R2 values of almost 0.64) demonstrate strong association of explanatory
variables and the outcome. Our main finding is that stronger state NAR membership and state
board make-up do have a significant effect on the likelihood that a state enacts minimum service
laws. However, the effects of both of these variables are in direct contrast to expectations set out
in the literature. A higher percentage of state NAR membership actually decreases the likelihood
that a state enacts minimum service laws. Similarly, a higher percentage of broker presence on a
state licensing board decreases the likelihood that a state enacts minimum service laws.
A state’s level of complaints filed against licenses and number of prelicensing hours are
also significant. As expected a higher state level of complaints filed against licenses increases the
likelihood that minimum service laws would be adopted. In contrast to expectations, a higher
level of prelicensing education hours increases the likelihood that minimum service laws would
be adopted.
Two economic variables, state population growth and income, are also significant
predictors of enactment. For protectionist reasons, we theorized that a state increase in any
economic variable would decrease the likelihood of law enactment. Our results, however, shows a
15
positive correlation. As state population growth increases, minimum service laws are more likely
to be adopted. As income increases, minimum services laws are also more likely to be adopted.
Our estimates indicate that continuing education requirements, consumer fund, change in
price, transactions, and GSP have no explanatory power for the variation and timing of the
enactment of minimum service laws.
6.1. With Time-Invariant Institutional Variables
We presented a hazard analysis framework where we model the law enactment process
conditioning on the adoption not yet having occurred. We start with standard discrete choice
models with time-invariant attributes. Exhibit 4 reports results from three different model
specifications – a simple linear probability model and non-linear logit and probit models. This
analysis is done with the state-level data (408 observations). We use the pre-adoption average
number of complaints per 100 licensees and association strength. Essentially, we assume that
these institutional characteristics are exclusive to the housing market. We still use the economic
variables as time-varying attributes since they are not directly associated with the institutional
characteristics of housing market.
The columns are distinguished by the assumption on the probability distribution. We also
allow the intercepts to differ across the census divisions. We start by estimating a simple linear
probability model (LPM), which is much simplified version of the adoption process. LPM
assumes that probability of adoption increases linearly with the dependent variables, implying a
constant marginal effect of dependent variables. This is particularly unrealistic in our study as
significant state level heterogeneity may exist. Moreover, out-of-range estimates may arise in
LPMs. Building on Column (1), we assume cumulative logistic function in Column (2), thus
avoiding the possibility of out-of-range estimates. In Column (3), we assume normal cumulative
16
distribution function i.e. probit model. The most important difference between logit and probit
model is that logistic function has slightly flatter tails. In our case, normal cumulative distribution
function i.e. the probit model may represent data better due to possibility of omitted variables in
model specifications.
Although significance varies across the columns, we generally find that a greater number
of complaints and stricter prelicensing educational requirement favor law enactment; and greater
association strength and industry influence do not support minimum services requirement.
Interestingly, democratic legislature control increases the likelihood of minimum service law
enactment.
Negative coefficients for percentage change in HPI and percentage change in transactions
seem to suggest that booming housing markets adversely affect the likelihood of requiring
minimum services. In a market with high appreciation rates and high transaction volume,
dissatisfaction by consumers goes unnoticed. However, as may be indicated by the positive
coefficients for GSP growth, population growth, and per capita income growth, it is also likely
that sound economic situation raises the number of instances where the broker services may fall
short of consumers’ expectation.
6.2. With Time-Varying Institutional Variables
Models in Exhibit 5 present further empirical results by including time-varying covariates
to address the measurement error problem and state level heterogeneity. Exhibit 5 takes the
specification, as in columns in Exhibit 4, and reports alternative econometric specifications. As
pointed out in Nanda (2008), incorporating time-varying covariates not only addresses the fact
that the law has been adopted in different years by different states, but also brings in more
information to the estimation system.
17
Duration dependence is assumed to follow standard Weibull distribution. Column (1) in
Exhibit 5 presents the baseline estimates from the hazard model with time-varying covariates
with no lagged values. We find significant estimates that are consistent with our postulates. As
assumed, association strength and industry influence put significant negative feedbacks to the
likelihood of enactment of the minimum services law. The more active housing market seems to
present positive feedback.
Legislators, however, may not have observed current year values in deciding on law
adoption. And, the law or the discussion around possible enactment might have affected the
current year observations, i.e. some endogenous feedbacks may persist. To address this concern,
we incorporate first lagged values of the institutional variables in column (2). Moreover, most
institutional variables tend to be measured with error. Legislators may want to consider ‘history’
or averages over a longer period to evaluate the need for a mandate. We take the average of first
to second lagged values for the institutional controls in column (3) to control for measurement
error. The estimates show remarkable improvement supporting the concerns of endogeneity and
measurement error. Significant estimates are consistent with our assumptions. As expected,
association strength and industry influence put significant negative feedbacks to the likelihood of
enactment of the minimum services law. The more active housing market seems to present
positive feedback.
The process of enacting minimum services requirement varies across states. Different
states in our sample may have different distributions for the duration dependence. In column (4),
we try to address this concern. We assume that the heterogeneity term  in equation (6) is
gamma distributed. However, we do not find any significant improvement over Column (3). xxvi
Since the institutional variables are measured with error, Column (3) presents our best estimates.
18
For interpreting the logit model in Column (2) of Exhibit 4, we estimate odds ratio, i.e.
measure of effect size, describing the strength of association between two binary data values.
Note that our estimation in Exhibit 5 follows a complimentary log-log specification, i.e. discrete
time proportional hazard model. The model includes logarithm of time as baseline hazard. The
beta (slope) coefficients estimated in the model are from the underlying proportional hazard
model. Exponentiated coefficient for the kth. regressor can be interpreted as a "hazard ratio" i.e.
rate at which spells will be completed at duration t; given that they survive until t. We report
odds ratio for the logit model in Column (2) of Exhibit 4 and hazard ratios for Column (3) of
Exhibit 5 to facilitate interpretation of our results in Exhibit 6. Four states – California, Montana,
South Carolina, and Wyoming - adopted minimum services requirement before 2000. To see
robustness of our results, we exclude those states and perform two key model analyses – Column
(2) Exhibit (4) and Column (3) Exhibit (5). Exhibit 7 presents the regression results from the
truncated sample. The results do not show any significant difference from what we find in
Exhibits 4 and 5.
In terms of magnitude, we calculate odds ratios and hazard ratios in Exhibit 6. Column
(1) of Exhibit 6 shows that an increase of one complaint per 100 licensees and each additional
hour of state prelicensing hours raise the odds of enacting a minimum services requirement by
factors of 1.80 and 1.54 respectively. And, a one percentage point increases in association
strength and industry influence raise the odds of NOT enacting a minimum services requirement
by factors of 0.21 and 0.79 respectively. In terms of hazard ratio from Column (2) of Exhibit 6, an
increase of one complaint per 100 licensees and each additional hour of state pre-licensing hours
increase the hazard rate of enacting a minimum services requirement by factors of 1.21 and 1.18
respectively. And, a one percentage point increases in association strength and industry influence
decrease the hazard rate of enacting a minimum services requirement by factors of 0.63 and 0.91
19
respectively. Alternatively, we can also interpret the percentage change effects of the attributes by
computing (hazard ratio - 1). An increase of one complaint per 100 licensees and each additional
hour of state pre-licensing hours increase the hazard rate of enacting a minimum services
requirement by 20.64 percent and 17.79 percent respectively. And, a one percentage point
increases in association strength and industry influence decrease the hazard rate of enacting a
minimum services requirement by 37.15 percent and 9.86 percent respectively.
7. Conclusion
This article has examined institutional and economic influences on state real estate
broker minimum service laws. The federal government and the academic literature have
speculated that the anti-competitive attitude of traditional brokers are the driving force behind the
enactment of these laws, and there is legislative evidence that state brokerage associations
introduced, supported, and/or lobbied for minimum service laws that passed. Our results,
however, show that factors reflecting strong state broker influence actually decreased the
likelihood that minimum services laws would be enacted in a state. While there may be many
traditional real estate brokers that are vocal against limited service brokerage, there may also be
many brokers that embrace the evolving nature of the brokerage industry brought about by
changing technology and new representational paradigms.
The stated purpose of many of the proposed minimum services bills that passed into law
was that of consumer protection. Our results indicate that rationale may be sincere, since the level
of state complaints against licensees is seen as the most significant indicator of enactment. This
suggests that states have made changes to broker licensing laws in an attempt to address state
perceived consumer protection concerns or problems.
20
This paper addressed the question of why minimum services laws were enacted.
Contrary to even our initial hypotheses, our results indicate that consumer protection related
factors - not anti-competition related factors - seem to be the driving force behind law enactment.
More research is now needed to determine the actual effect of enacted minimum service laws on
both broker competition and on consumer protection.
References
Association of Real Estate Licensing Law Officials (ARELLO) Digest of Real Estate Licensing
Laws and Current Issues (1999-2007).
de Figueiredo Jr., Rui J.P., and R.G. Vanden Bergh. “The Political Economy of State-Level
Administrative Procedure Acts” The Journal of Law and Economics Vol. XLVII (2004).
Federal Trade Commission and U.S. Department of Justice, Competition in the Real Estate
Brokerage Industry, Report by the Federal Trade Commission and U.S. Department of Justice,
April 2007.
Government Accountability Office, Real Estate Brokerage: Various Factors May Affect Price
Competition, Report to the Subcommittee on Housing and Community Opportunity, Committee
on Financial Services, House of Representatives, #GAO-06-1005T, July 2006.
Hahn, R.W., R. E. Litan, and J. Gurman, Bringing More Competition to Real Estate Brokerage,
Real Estate Law Journal, 2006, 35, 86-118. (This was originally a working paper Paying Less for
Real Estate Brokerage: What Can Make it Happen?, AEI Brookings Joint Center for Regulatory
Studies, Working Paper 05-11, 2005).
Kiefer, N.M. Economic Duration Data and Hazard Functions, Journal of Economic Literature
Vol. 26, No. 2 (1988), 646-769.
21
Levitt, S. and C. Syverson, Antitrust Implications of Home Sellers Outcomes When Using FlatFee Real Estate Agents, Brookings-Wharton Papers on Urban Affairs 2008, 2008, 47-93.
Magura, Matthew, How Rebate Bans, Discriminatory MLS Listing Policies, and Minimum
Service Requirements Can Reduce Competition for Real Estate Brokerage Services and Why it
Matters, U.S. Department of Justice - Economic Analysis Group, May 2007.
Miceli, T. J ., K.A. Pancak, and C. F. Sirmans. “Is the Compensation Model for Real Estate
Brokers Obsolete? The Journal of Real Estate Finance and Economics Vol. 35, Issue 1 (2007)
Nanda, A. Property Condition Disclosure Law: Why Did States Mandate ‘Seller Tell All’?, The
Journal of Real Estate Finance and Economics Vol. 37, Issue 2 (2008).
Pancak, K.A., A Critical Examination of Broker Minimum Service Laws, Working Paper,
University of Connecticut Center for Real Estate 2008.
White. Lawrence, The Residential Real Estate Brokerage Industry: What Would More Vigorous
Competition Look Like?, Stern School of Business, New York University, revised draft April 6,
2006.
Woodall, Peter and Stephen Brobeck, “State Real Estate Regulation: Industry Dominance and Its
Consumer Costs,” Consumer Federation of America (July 2006).
Wooldridge, J. Econometric Analysis of Cross-Section and Panel Data, MIT Press (2002).
22
EXHIBIT 1: STATES WITH NON-WAIVABLE MINIMUM SERVICE REQUIREMENTS
STATE
YEAR
SECTION
ENACTED
Enacted after 2000
Alabama
2005
Arizona
2005
Year language
about taking
reasonable
steps to assist
a client in
confirming
information
was added.
CODE OR
REGULATION
SECTION
Alabama Code §§
34-27-84 (c)
Arizona
Administrative Code
R4-28-1101
SUMMARY OF
APPLICABLE PROVISION(S)
At a minimum, all listing brokers must:
- accept delivery of and present all offers to assist
the consumer in negotiating offers
- answer the consumer's questions relating to the
transaction
Brokers must do the following for a client:
- use reasonable care to obtain information
material to a client's interests and relevant to the
contemplated transaction, and communicate the
information to the client
- take reasonable steps to assist a client
- take reasonable steps to assist a client in
confirming the accuracy of information relevant
to the transaction
Licensees must perform acts expeditiously, and can’t
intentionally or negligently delay performance
Idaho
2007
Idaho Statute § 542087(3)
Illinois
2004
225 Illinois
Compiled Statutes
454, Article 15,
Section 75
Indiana
2006
Indiana Code 2534.1-10-9.5
If a broker enters into a written contract for
representation to represent a client, the broker must:
- be available to the client to receive and timely
present offers.
This duty is mandatory and can’t be waived.
While the state allows non-agency, this requirement
applies to all types of representation or customer
service agreements.
In an exclusive brokerage agreement, the broker
must provide the following services:
- accept delivery of and present offers
- assist the client in developing, communicating,
negotiating, and presenting offers
- answer client questions
If a broker DOES NOT have an agency
relationship with a consumer, at a minimum the
broker has to perform the following:
- be available to receive and present offers
- assist in negotiating, completing real estate
forms, communicating
- respond to questions
If a second broker performs those duties for the
consumer, because the first broker failed to perform
them, there would not be an agency relationship
between the second broker and the consumer.
If a broker DOES have an agency relationship
with a client, the broker must fulfill the terms of the
agency relationship, and present all offers
immediately upon receipt.
23
Iowa
2005
Iowa Code §
543B.56A
Kentucky
2005
201 Kentucky
Administrative
Regulations 11:045
Missouri
2005
Missouri Revised
Statute § 339.780 (7)
Texas
2005
Texas Occupations
Code Title 7
§1101.557
Utah
2005
61-2-27 Utah Code
Annotated
Enacted before 2000
California
1987
California
Civil Code
Section 2079
Montana
1995
Montana Code
Annotated § 37-51313 (12)
South
before 2000
South Carolina Code
of Laws Section 40-
Only applies to non-agency situations. Does not
apply when a licensee represents a client in a
transaction, but only when a licensee enters into a
written agreement that does not involve agency
At a minimum, all brokerage agreements must
state that the broker will:
- accept delivery of and present offers
- assist the client in developing, communicating,
negotiating, and presenting offers
- answer the client's questions relating to the
brokerage agreements and negotiations
- provide prospective buyers access to listed
properties
At a minimum, all brokers representing a client
must:
- accept delivery of and present all offers
- accept all earnest money deposits that are
presented to the broker
- assist clients in developing, communicating,
negotiating, and presenting offers
- answer questions relating to offers
Failure to comply with these minimum requirements
is considered gross negligence.
Regulations limits choice; interesting that bill
proposing minimum services was defeated in 2006.
Allows non-agency – transaction broker
In an exclusive brokerage agreement, the broker
must provide the following services:
- accept delivery of and present offers
- assist the client in developing, communicating,
negotiating, and presenting offers
- answer client questions
In an exclusive brokerage agreement, the broker
must provide the following services:
- present offers to and from client
- answer client questions
In an exclusive brokerage agreement, the broker
must provide the following services:
- accept delivery of and present offers
- assist the client in developing, communicating,
negotiating, and presenting offers
- answer client questions
Listings brokers must conduct a reasonably
competent and diligent visual inspection of listed
property, and disclose all facts materially affecting
the value or desirability of the property that that the
inspection revealed.
Licensees must “endeavor to ascertain all pertinent
facts concerning each property in any transaction in
which the licensee acts” so the licensee can fulfill the
his or her obligation to avoid error, exaggeration,
misrepresentation, or concealment of pertinent facts
Licensees must:
- upon receipt, prepare and present offers
24
Carolina
Wyoming
57-135 (D) (1)
before 2000
Wyoming Code §
33-28-111 (xxix)
-
deliver written acceptances of offers to all
parties
- ensure that all of the terms and conditions of the
transaction are included in the offer
- ensure that changes or modifications made
during negotiation are in writing and initialed
and dated by both parties before proceeding with
the transaction.
Licensees must:
- advise buyer and seller of all terms of a
proposed sale at the time an offer is presented
including estimated discounts and closing costs
- submit all offers to a seller
25
EXHIBIT 2: INDEPENDENT VARIABLES
Variables
Definition
Institutional Variables
COMPLAINTS
Number of Complaints per 100 Licensees
NAR
State National Association of REALTORS, Inc. membershipxxvii as a
percentage of active state licenseesxxviii
BOARD
Percentage of real estate industry members on the state licensing
boardxxix
EDUCATION
Pre-Licensing Hours Required for Salespersonsxxx
CE EDUCATION
Continuing Education Hours Requirement
CONSUMERFUND
Maximum consumer protection fund liability payout against a broker
(in thousands of dollars)xxxi
POLITICS
Binary Variable indicating whether a state legislature is controlled by
democrats or others (including republican)xxxii
Economic Variables
PRICE
Average of Percent YoY Change in Quarterly OFHEO Purchase-Only
State House Price Index
TRANSACTIONS
Percent Change in Number of transactions
GSP
Percent Change in Real Gross State Product
POPULATION
Percent Change in Population
INCOME
Percent Change in Per Capita Income
26
EXHIBIT 3: SUMMARY STATISTICS
50 States Characteristics: 2000—2007:
408 Observations
Variables
N
Mean
Std. Dev.
COMPLAINTS
395
1.856
12.187
NAR
393
50.688
15.435
BOARD
408
69.243
18.696
EDUCATION
408
58.921
33.004
CE EDUCATION
400
6.440
6.188
CONSUMERFUND
408
13509.80
16988.36
POLITICS
408
0.367
0.483
PRICE
408
7.031
5.150
TRANSACTIONS
402
2.656
10.630
GSP
408
2.464
2.087
POPULATION
408
0.908
0.866
INCOME
408
4.380
2.805
NOTES: Missing information is the reason behind less than 408 observations for some
variables.
27
EXHIBIT 4: STANDARD DISCRETE CHOICE MODELS WITH TIME-INVARIANT
INSTITUTIONAL CONTROLS
(Dependent Variable: Law Adoption Dummy)
Regressors
(1)
(2)
(3)
0.0033
(0.0031)
0.5881
(0.4170)
0.2998***
(0.1796)
-0.0028**
(0.0013)
-1.5596*
(0.4894)
-0.9009*
(0.2517)
BOARD
-0.0009
(0.0012)
-0.2333**
(0.0949)
-0.1279*
(0.0464)
EDUCATION
0.0013**
(0.0005)
0.4318*
(0.1369)
0.2475*
(0.0675)
CE EDUCATION
-0.0029
(0.0033)
-0.1142
(0.1228)
-0.0467
(0.0629)
CONSUMERFUND
-0.0000
(0.0000)
0.0000
(0.0000)
0.0000
(0.0000)
POLITICS
0.0484
(0.0542)
20.5719*
(6.2052)
11.8353*
(2.8881)
PRICE
0.0034
(0.0032)
-0.1498
(0.1037)
-0.0799
(0.0628)
TRANSACTIONS
-0.0027
(0.0016)
-0.0569
(0.0371)
-0.0336
(0.0216)
GSP
-0.0008
(0.0061)
0.0912
(0.2490)
0.0286
(0.1427)
POPULATION
0.0771**
(0.0314)
8.5707*
(2.4075)
4.9443*\
(1.2904)
INCOME
0.0079***
(0.0043)
0.1126
(0.2026)
0.0846
(0.1023)
Census
Division
Linear
Probability
Model
Regression
0.175
Census Division
Census Division
Logit Regression
Probit Regression
COMPLAINTS
NAR
Fixed Effect?
Model
R2/Pseudo R2
0.643
0.645
Log Likelihood
-30.906
-30.679
N
394
258
258
NOTES: Clustered Standard errors are reported within parentheses. ‘*’, ‘**’, and
‘***’ denote 1 percent, 5 percent and 10 percent significance level. This analysis
includes data for all the states, from 2000 to 2007. The United States is divided
into nine census divisions as defined by US Census Bureau e.g. New England.
28
EXHIBIT 5: PROPORTIONAL HAZARD MODELS WITH TIME-VARYING
INSTITUTIONAL CONTROLS
(Dependent Variable: Law Adoption Dummy)
Regressors
(1)
(2)
(3)
(4)
COMPLAINTS
-0.1019
(0.2263)
0.0726
(0.0938)
0.1876***
(0.1108)
-0.5716
(0.5724)
NAR
-0.2191*
(0.0738)
-0.2992***
(0.1582)
-0.4644**
(0.2278)
-0.0106
(0.0205)
BOARD
-0.0882**
(0.0367)
-0.1036
(0.0687)
-0.1038***
(0.0549)
-0.0067
(0.0166)
EDUCATION
0.1034*
(0.0269)
0.1274**
(0.0559)
0.1638**
(0.0684)
0.0288
(0.0189)
CE EDUCATION
-0.2902
(0.1873)
-0.4099
(0.3712)
-0.3678
(0.3022)
-0.0847
(0.0857)
CONSUMERFUND
0.0001
(0.0001)
0.0001
(0.0001)
0.0001
(0.0001)
-0.0001
(0.0001)
POLITICS
6.3178*
(1.4139)
8.5929**
(3.9544)
9.5454**
(4.3553)
0.1932
(0.8799)
PRICE
0.2067
(0.1844)
0.1254
(0.0885)
0.4039**
(0.1949)
-0.1596
(0.1097)
TRANSACTIONS
0.1392**
(0.0641)
0.1587***
(0.0939)
0.0859
(0.0647)
-0.0498
(0.0476)
GSP
-0.4084
(0.3316)
-0.7128
(0.5258)
-1.4693***
(0.8787)
0.0577
(0.4239)
POPULATION
4.4288**
(2.1664)
7.9529
(4.9417)
7.6349***
(4.6287)
0.7133
(0.9020)
INCOME
-0.3089**
(0.1253)
-0.7107
(0.4401)
-0.5921***
(0.3159)
0.1859
(0.3419)
Model Description
Current
TimeVarying
Attributes
First Lagged
Time-Varying
Attributes
Modeling Concern
More
information
Regulators do
not observe
current values
Avg. of 1st.
and 2nd.
Lagged TimeVarying
Attributes
State-level
Heterogeneity
Weibull
Weibull
Avg. of 1st.
and 2nd.
Lagged TimeVarying
Attributes
Institutional
Variables
measured with
error
Weibull
Distribution for Duration
Dependence
Distribution for Heterogeneity Term
Weibull
Gamma
Log Likelihood
-20.854
-18.768
-14.444
-30.991
N
248
223
192
294
NOTES: Models include logarithm of time as the baseline hazard specification. Clustered standard errors are
reported within parentheses. ‘*’, ‘**’, and ‘***’ denote 1 percent, 5 percent and 10 percent significance levels.
This analysis includes data for all 50 states, from 2000 to 2007. We used time-invariant institutional variable in
table 2. However, in this table, we use time-varying institutional controls.
29
EXHIBIT 6: ODDS RATIO AND HAZARD RATIO
(Dependent Variable: Law Adoption Dummy)
Regressors
Column (2), Exhibit (4)
Odds Ratio
1.8006
Column (3), Exhibit (5)
Hazard Ratio
1.2064***
NAR
0.2102*
0.6285**
BOARD
0.7918**
0.9014***
EDUCATION
1.5401*
1.1779**
CE EDUCATION
0.8921
0.6922
CONSUMERFUND
1.0001
1.0001
8.60e+08*
13980.61**
PRICE
0.8608
1.4977**
TRANSACTIONS
0.9447
1.0897
GSP
1.0955
0.2301***
5275.257*
2069.325***
1.1192
0.5531***
COMPLAINTS
POLITICS
POPULATION
INCOME
30
EXHIBIT 7: LOGIT AND PROPORTIONAL HAZARD MODELS
(EXCLUDING CA, MT, SC, AND WY)
(Dependent Variable: Law Adoption Dummy)
Regressors
Column (2)
Exhibit (4)
0.5881
(0.4174)
Column (3)
Exhibit (5)
0.1876***
(0.1109)
NAR
-1.5596*
(0.4899)
-0.4644**
(0.2281)
BOARD
-0.2333**
(0.0949)
-0.1038***
(0.0550)
EDUCATION
0.4318*
(0.1371)
0.1638**
(0.0685)
CE EDUCATION
-0.1142
(0.1230)
-0.3678
(0.3025)
CONSUMERFUND
0.0001
(0.0001)
0.0001
(0.0001)
POLITICS
20.5719*
(6.2114)
9.5451**
(4.3597)
PRICE
-0.1498
(0.1037)
0.4039**
(0.1951)
TRANSACTIONS
-0.0568
(0.0371)
0.0859
(0.0648)
GSP
0.0912
(0.2492)
-1.4692***
(0.8795)
POPULATION
8.5707*
(2.4099)
7.6348***
(4.6333)
INCOME
0.1126
(0.2027)
-0.5921***
(0.3162)
Logistic Regression
Avg. of 1st. and 2nd.
Lagged TimeVarying
Attributes
COMPLAINTS
Model Description
Modeling Concern
Variables measured
with error
Distribution for Duration
Weibull
Dependence
Log Likelihood
-30.906
-14.444
N
242
180
NOTES: Models include logarithm of time as the baseline hazard specification.
Clustered standard errors are reported within parentheses. *’, ‘**’, and ‘***’ denote
1 percent, 5 percent and 10 percent significance levels. This analysis includes data
for all 50 states, from 2000 to 2007. We used time-invariant institutional variable in
table 2. However, in this table, we use time-varying institutional controls in sample
without California, Montana, South Carolina, and Wyoming.
31
i
In most states, a “broker” enters into an agency agreement with a client, and a “salesperson”
works for a broker. Both brokers and salespersons that represent a client are “agents” of the
client. Both also need to be licensed by their state to engage in the real estate business, and if
licensed are also referred to as “licensees”. Throughout this paper, the term broker will be used.
ii
It is important to make the distinction between minimum service laws that increase the
brokerage services a consumer must purchase as opposed to laws that define the special agency
relationship between a broker and client. Many states have provisions codifying common law
agency fiduciary duties of obedience, loyalty, disclosure, confidentiality, accounting, and
reasonable care; these types of laws are not considered minimum service laws.
iii
The DOJ list can be found at http://www.usdoj.gov/atr/public/real_estate/fee_details.htm . As
of January 8, 2009, the DOJ listed the following states have limiting choice due to minimum
service requirements: Alabama, District of Columbia, Idaho, Illinois, Indiana, Iowa, Kansas,
Maine, Missouri, Oregon, South Dakota, Texas, Utah, Washington, and West Virginia.
iv
National Association of Realtors®, Inc. Research Division, Structure, Conduct, and
Performance of the Real Estate Brokerage Industry, November 2005.
v
Roberts, Glenn, Inman News, Federal antitrust agencies attack Missouri real estate law,
May 24, 2005, quoting Laurie Janik, general counsel for the National Association of Realtors,
April 22, 2005 letter to state Realtor association executives
vi
http://www.usdoj.gov/atr/public/real_estate/fee_details.htm.
vii
This was decided in the U.S. Supreme Court case of Parker v. Brown, 317 U.S. 342 (1943).
However, the DOJ has sued the National Association of Realtors for the alleged anti-competitive
nature of the Associations Virtual Office Website (VOW) policy. The Association had adopted a
policy permitted brokers to selectively withhold or “opt-out” of allowing other brokers to
advertise MLS listing information online. The DOJ and Association settled this lawsuit in 2008.
viii
Illinois Senate Transcript, 3/24/2004, found at
http://12.43.67.2/senate/transcripts/strans93/09300090.pdf
ix
Illinois Association of Realtors®, Provisions of Senate Bill 2887 (Public Act 93-957), revised
10-04, at http://www.illinoisrealtor.org/Member/government/issues/sb2887.asp .
x
“The Value of Full Service”, Chicago Agent, Vol. 2, Issue 2 (2005), at
http://www.chicagoagentmagazine.com/issue/article.asp?id=4&article=186&status=archived
xi
Blanche Evans, Remember the Alamo: Limited Service Dust-up Brings Out State's Rights
Argument Against DOJ, FTC, REALTY TIMES, Apr. 22, 2005;
http://realtytimes.com/rtcpages/20050422_dojstepsin.htm
xii
Alabama Real Estate Commission Update Newsletter, Fall 2005,
http://www.arec.state.al.us/pdf/Update/arec_update_rev1011.pdf.
xiii
Iowa House File 375, An Act Relating To The Duties Imposed On A Real Estate Broker By A
Brokerage Agreement (unanimously passed both the Iowa House and Senate).
xiv
Iowa Real Estate Commission Minutes, April 14, 2005,
http://www.state.ia.us/government/com/prof/sales/meet_schedule/2005/MIN4_14_05.htm
xv
The Arizona Department of Real Estate Bulletin, May 2005, Volume 31, Issue 2;
http://www.re.state.az.us/PUBLIC_INFO/Bulletins/Bulletin_05.05.pdf
xvi
Kit Wagar, Real Estate Brokers Face New Restrictions in Missouri, July 14, 2005, The
Kansas City Star, as reported online by RISmedia at http://rismedia.com/2005-07-14/real-estatebrokers-face-new-restrictions-in-missouri/
xvii
Kentucky Senate Bill 43, 2006 Session, www.lrc.ky.gov/record/06RS/SB43.htm. “Kentucky
Shoots Down Real Estate Minimum-Service Proposal” Glenn Roberts Jr., Inman News, February
17, 2006, http://beta.inman.com/news/2006/02/5/kentucky-shoots-down-real-estate-minimumservice-proposal.
xviii
See Pancak (2008).
32
xix
Idaho Real Estate Commission Guideline #23, effective July 1, 2007, adopted January 17,
2008; http://www.idahorealestatecommission.com/guidelines/guideline23.pdf.
xx
Idaho House Bill 135 (2007).
xxi
Additionally, seven states enacted laws since 2004 providing for minimum services: Florida,
Michgan, Nevada, New Mexico, Ohio, Tennessee, Wisconsin (see Pancak 2008) The laws in
those states, however, are not mandatory in that they allow a consumer to waive the services. For
purposes of our analysis, however, we will only include states that enacted non-waivable
minimum service laws after 2000.
xxii
For example, California’s law requiring that listing brokers conduct a reasonable competent
and diligent inspection of listed property was in response to a California court case related to
property condition disclosre.
xxiii
The agencies also discussed other motivations for the laws.
xxiv
Woodall and Brobeck (2006).
xxv
The baseline specification draws on Kiefer (1988) and Wooldridge (2002).
xxvi
Due to many small omitted influences on law adoption, normal distribution is expected to
represent data better than gamma distribution.
xxvii
Data on NAR membership from National Association of REALTORS®, Inc. monthly
membership report for years ending December 31, 1999 - 2007.
xxviii
ARELLO (1999-2007), “SALESPERSONS: Active Salespersons” and “BROKERS: Active
Brokers” plus “BROKERS: Active Associate Brokers”.
xxix
ARELLO (1999-2007), “# members” heading and “# Industry Members” heading.
xxx
ARELLO (1999-2007), “HOURS PRE-” heading.
xxxi
ARELLO (1999-2007), “MAX. FUND Liability: Broker” heading.
xxxii
Data compiled from the National Conference of State Legislatures (NCSL).
33