Advertisement

Eastern Economic Journal

, Volume 44, Issue 1, pp 69–83 | Cite as

Bringing the Effects of Occupational Licensing into Focus: Optician Licensing in the United States

  • Edward J Timmons
  • Anna Mills
Original Article

Abstract

The labor market institution of occupational licensing continues to grow in scope in the United States and abroad. In this paper, we estimate the effects of occupational licensing on opticians using data from the US Census and American Community Survey. Our results suggest that optician licensing is associated with opticians receiving as much as 16.9 percent more in annual earnings. In an examination of malpractice insurance premiums in all states and participation rates in optician certification programs in Texas, we find little evidence that optician licensing has enhanced the quality of services delivered to consumers. By and large, optician licensing appears to be reducing consumer welfare by raising the earnings of opticians without enhancing the quality of services delivered to consumers.

Keywords

occupational licensing occupational regulation 

Keywords

J44 I18 

As of 2006, occupational licensing affected 29 percent of the workforce in the United States, and the percentage of the US workforce directly affected by occupational licensing continues to grow [Kleiner and Krueger 2013]. Generally, an occupation can be regulated in three ways: registration, certification, and licensing. Registration requires individuals to provide some level of information to a government agency, such as their names, addresses, and qualifications. Once they have provided the information to the government, they can begin practicing. The second form of regulation, certification, restricts practitioners from using a professional title. For instance, only individuals who have passed an examination and met additional criteria may use the title “certified financial analyst.” The most stringent form of occupational regulation is licensing. Licensing requires any individual who wishes to practice to meet specific standards set by the government [Rottenberg 1962]. Certification systems allow uncertified professionals to practice, but licensing systems do not permit unlicensed professionals to practice.

This study specifically examines the effects of occupational licensing on opticians. Opticians are licensed in 21 states in the United States, and Texas requires certification. Opticians have many responsibilities, which include interpreting the prescriptions from optometrists and ophthalmologists, collecting eye measurements, helping individuals select contact lenses and eyeglasses, and ensuring that eyeglasses are adjusted properly [Optician Training 2014]. The decision to examine the optician market is partially motivated by the report License to Work by the Institute for Justice [Carpenter et al. 2012]. In this report, the Institute for Justice ranks opticians as having the 6th most burdensome requirements and the 42nd most heavily regulated occupation of the 102 low- and moderate-income occupations studied. The requirements vary from state to state, but the average requirements for states that license are three exams, 2 years of education, and US$184 in registration fees [Carpenter et al. 2012].

To estimate the effect of licensing on optician earnings, we pool cross-sectional data from the 1940–2000 Census and American Community Survey data from 2001 to 2012. During this time period, 17 of the 21 currently licensing opticians passed licensing statutes and a handful of states adjusted licensing requirements. We exploit these states changing status to employ a difference-in-difference framework to estimate the effects of the laws on practitioner earnings.

After a summary of some of the existing literature on the theory and empirical estimates of the effects of licensing on low-income professions, we estimate how state licensing of opticians has affected opticians’ earnings. Our results suggest that optician licensing is generally associated with higher optician earnings.

LITERATURE REVIEW

Economic theory paints two different pictures of the effects of occupational licensing. Supporters of occupational licensing believe that it protects consumers by improving the quality of service. Occupational licensing has become particularly relevant as the US economy has shifted from manufacturing to service industries. Measuring quality performance is more difficult in service industries, and thus, licensing has become the main method of showing competency in an occupation [Kleiner 2000]. According to Kenneth Arrow [1971], occupational licensing has the potential to minimize consumer uncertainty, and therefore lead to an increase in overall demand for the service. Leland [1979] argues that requiring a minimum level of training produces positive social payoffs and reduces the asymmetric nature of the market. Thus, occupational licensing is believed to be necessary to promote the public interest of safety and to ensure that the services rendered meet minimum quality standards.

Several economists, however, are skeptical of the benefits of occupational licensing. Smith [(1776) 1994] believed it to be a way to “limit the number of apprentices per master, thus ensuring higher earnings for persons in these occupations.” Freidman [1962] questioned whether the government and professional organizations were “unbiased gatekeepers” and whether the professional organizations were establishing monopoly rents by creating more difficult barriers to entry, thereby restricting the supply of practitioners and resulting in higher professional earnings. Minimum quality standards may become both a floor and a ceiling as declining competition leads to less incentive to innovate and improve [Kleiner 2006]. Another side effect of licensing is that it promotes the idea among practitioners that higher quality will result only if a higher wage is guaranteed [Gellhorn 1976]. Shapiro [1986] points out that licensing imposes certain standards that pass on to all consumers, despite clear differences in consumer valuation of the quality of the service. Thus, imposing quality standards does not guarantee a positive experience for all consumers. In a related study, Maurizi [1974] notes that as the demand for an occupation grew, the pass rate on licensure exams for the occupation fell. Furthermore, in a nationwide study of the effects of occupational licensing, Kleiner and Krueger [2010] find evidence that licensing increases wages by 15 percent. This result demonstrates the magnitude of the gains to practitioners from policies that create barriers to entry into a profession.

Several studies have attempted to estimate the economic effects of occupational licensing. This study focuses on those occupations that do not require a substantial amount of training, and those in which practitioners receive low or moderate levels of pay. Existing studies have estimated the effect of cosmetology regulation on prices of services [Adams et al. 2002], as well as the effect of English proficiency requirements for Vietnamese workers obtaining a license to enter the manicurist profession [Federman et al. 2006]. In a study analyzing the licensing of barbers, Timmons and Thornton [2010] show that tougher licensing requirements increased earnings between 11 and 22 percent. They also find evidence that reductions in the supply of barbers were the primary mechanism for the wage increase. Barbering is an occupation that is licensed throughout the United States. In a separate analysis of an occupation that is not universally licensed and has only recently become subject to regulation, Thornton and Timmons [2013] find evidence that licensing increases massage therapist earnings by 16.2 percent. In the small number of states that certify the massage profession, evidence of a similar earnings premium was less convincing. This finding also suggests that licensing generally increases rents rather than the quality of the service provided to consumers. In another study, examining the effects of licensing of radiologic technologists — an occupation that requires relatively low amounts of training and is not licensed in all states — Timmons and Thornton [2008] find that licensing increases the earnings of radiologic technologists by 3.3–6.9 percent. Despite the relatively low barriers to entry of each of these occupations (compared with physicians, for example), there is still evidence of substantial economic effects from occupational licensing. The magnitude of the estimated effects differs — perhaps because of the employment arrangement. Massage therapists and barbers are likely to have more autonomy than radiologic technologists. Other researchers have speculated that the potential for occupational licensing to result in large economic rents for practitioners may depend on the degree of autonomy the professional enjoys (that is, whether the professional is an employee or is self-employed) [Kleiner 2006].

One issue receiving considerably less attention in the existing literature is how the effect of licensing on the earnings of professionals varies over time (what might be referred to as the “duration effect”). Thornton and Timmons [2013] explore this issue in the MT profession and find that licensing has the largest effect on wages after it has been in effect for 10 years. A working paper by Han and Kleiner [2015] finds a similar result.

Papers testing the effects of occupational licensing on the quality of services delivered to consumers are more mixed and smaller in number primarily because of difficulties in assessing the quality of service delivered to consumers. A recent report from The White House [2015] summarizes studies that have attempted to measure the impact of licensing on quality and most find little evidence of a measurable effect. Perhaps this result is not that surprising considering that professional associations (as opposed to consumers) are the fiercest supporters of new and existing occupational licensing laws [Kleiner 2013]. By examining an occupational that has been licensed in some states for more than 80 years, our study attempts to estimate the effect of licensing duration. We also look for evidence that optician licensing has enhanced the quality of services delivered to consumers. Before discussing the effects of licensing on opticians, we will provide a brief overview of the profession.

REGULATION OF THE VISION CARE MARKET

The market for vision care is divided into three groups: ophthalmologists, optometrists, and opticians. Ophthalmologists are medical doctors who can diagnose and treat eye diseases, as well as perform eye examinations to prescribe and dispense contact lenses, and eyeglasses. Optometrists are not medical doctors; they are health-care professionals who can also perform eye examinations to prescribe and dispense contact lenses, and eyeglasses. Opticians dispense eyeglasses and contact lenses, and do not have the authority to diagnose or treat eye diseases or perform eye examinations. Generally, all three professions have supported expanding licensing of the optician profession. As with nearly all licensing statutes, professional associations (in this specific case, the Opticians Association of America) are the primary catalyst for licensing legislation. State optician groups have historically lobbied for licensing on the grounds that it would both signal quality to consumers and restrict entry to the profession [Haas-Wilson and Savoca 1990]. Consumers may feel more comfortable purchasing eyeglasses and contact lenses from a licensed practitioner — if one assumes that consumers are aware of licensing legislation.

Regardless of the outcome (restricted competition or perceived higher quality), consumers would be forced to pay higher prices for eyeglasses. Ophthalmologists and optometrists have also historically supported optician licensing, but purely on the grounds of limiting competition and protecting market share [Maurizi et al. 1981].1 Ophthalmologists and optometrists fear that unregulated opticians may be able to offer eyeglasses at substantially lower prices. As a further effort to control competition, ophthalmologists and optometrists have tried to limit optician autonomy. Twenty-two states have passed laws that indirectly affect opticians’ ability to be independent rather than tied to an ophthalmologist or optometrist. Four of those states do not allow opticians to fit contact lenses, and 16 of them require either an ophthalmologist or an optometrist to be present when fitting lenses.

Why have efforts to license opticians been slow to emerge or renew? Growing competition from online and mail order outlets (1–800 Contacts, for example) may have led to substantial increases in competition, particularly in the contact lens market. In 2004, the Fairness to Contact Lens Consumers Act was signed into law by then president George W. Bush. The law required professionals to provide prescription details to their patients, but also gave states the right to opt out of the law and set their own guidelines (for instance, allowing professionals to refuse to provide prescription details if the prescriptions are more than 1 year old) [Tedeschi 2004]. Lobbying efforts on the part of professional associations have primarily focused on thwarting efforts by 1–800 Contacts to roll back individual state guidelines on refusing access to prescription information [Tedeschi 2004].

The Federal Trade Commission conducted a study to observe the quality of eye care provided by licensed versus unlicensed professionals, and found that the quality difference between licensed and unlicensed professionals was statistically insignificant [Cox and Foster 1988]. States with bans on optometrist and optician price advertising also have been shown to have significantly higher prices — as much as 16 percent more than states without similar bans [Feldman and Begun 1978]. As noted earlier, licensing may serve as a signal to consumers that practitioners have met minimum quality standards. But regulation in the vision care market does not always benefit the consumer. Empirical studies have shown that as the level of professional control increases, such as a requirement for supervision of opticians, the price of eyeglasses increases [Benham and Benham 1975]. These specific examples show that the interests of consumers are not always represented by regulatory intervention in the vision market.

DATA AND PRELIMINARY ANALYSIS

Correlation between optician regulation and earnings

What is the scope of regulation in the optician market today? Table 1 presents specific information on the states that regulate opticians. Figure 1 depicts the states with licensing. Data were gathered from License to Work, and then confirmed by consulting the annotated statutes and licensing boards of each state [Carpenter et al. 2012]. All the states listed in Table 1 require opticians to be licensed (excluding Texas, which has a certification law). South Carolina was the first state to require licensing of opticians, in 1917. Connecticut, New York, and Rhode Island passed legislation in the 1930s. A second wave of states (12) adopted licensing legislation between 1949 and 1957. Alaska and Vermont both began to license the profession in 1973, followed by Ohio and Arkansas in 1980 and 1981, respectively. The final state to adopt licensure was California in 1988.
Table 1

State Regulation of Opticians

State

Year licensing was introduced

One-time fee ($)

Experience or education (days)

Number of exams

Minimum years of education

Age (years)

Alaskaa

1973

275

420

2

12

0

Arizona

1956

200

1,095

3

12

0

Arkansas

1981

200

1,120

2

12

21

California

1988

141

0

2

12

18

Connecticut

1935

100

730

4

0

0

Florida

1949

850

730

3

12

18

Georgia

1956

115

700

3

12

18

Hawaii

1949

75

700

2

12

18

Kentucky

1954

50

730

3

12

18

Massachusetts

1955

54

730

3

0

0

Nevada

1951

350

1,128

3

12

18

New Jersey

1952

25

857

2

12

0

New Yorkb

1936

100

560

3

12

18

North Carolina

1951

250

910

1

12

18

Ohio

1980

96

467

2

12

18

Rhode Islandc

1937

70

1,095

2

12

18

South Carolina

1917

150

730

3

12

0

Tennessee

1955

278

730

3

12

18

Texas (Certification)

1976

105

1

2

0

0

Vermont

1973

70

730

1

12

18

Virginia

1950

300

730

3

12

18

Washington

1957

200

731

3

12

18

aDropped education requirement from 1,400 days to 420 days in 2002.

bAdded an exam in 1973.

Increased education requirements from 365 days to 1,095 days in 1974.

Sources: Carpenter et al. [2012], and each state’s licensing board and licensing statutes.

Figure 1

State Regulation of Opticians.

Source: Carpenter et al. [2012], and each state’s licensing board and licensing statutes.

The licensing requirements of the states that require licensure differ substantially. The fees for obtaining an optician license vary from $70 in Vermont to as much as $850 in Florida. Education and experience requirements are as little as no education (beyond compulsory levels of schooling) in California to as much as 1,128 days in Nevada. Every state requires an exam, but the number of exams required varies from only one in North Carolina and Vermont to four in Connecticut. Many states also specify whether applicants are required to complete high school or be a minimum age.

By studying annotated statutes, we identify a handful of states that enacted changes in licensing requirements. In 2002, Alaska substantially reduced optician licensure requirements from 1,400 days of education, and experience to 420. New York and Rhode Island, in contrast, made existing requirements stricter: New York added an additional exam in 1973, and Rhode Island substantially increased education and experience requirements from 365 days to 1,095 days.

What are the economic effects of optician licensing? We obtained data from the US Census for 1940–2000 and the American Community Survey for 2001–2012 [Ruggles et al. 2010]. The focus is on individuals identifying themselves as opticians and reporting annual earnings above 0.2 We identify states with licensing by comparing the date the licensing law was passed, and the year of the observation. If the statute was passed 1 year before the survey year, we classify the observation as “licensed.” For example, we classify observations from the 1950 Census (the survey was conducted in 1949) in Virginia as “not licensed,” but Virginia observations from 1960 to the present are classified as “licensed.” Table 2 contains a simple comparison of states with and without licensing statutes over the sample period. Texas is excluded from the comparison because it has a certification law.
Table 2

Summary Statistics of the 1940–2012 US Census and American Community Survey Optician Sample

Item

States without optician licensing(% except where otherwise noted)

States with optician licensing(% except where otherwise noted)

 

Mean

Median

Mean

Median

Annual earnings ($2012)

$29,765

$26,316

$36,782

$33,316

Age (years)

38.4

37.0

40.1

39.0

Male

37.8

 

43.6

 

African American

3.9

 

4.5

 

Other minority

4.1

 

9.0

 

Hispanic

1.6

 

5.4

 

Associate degree

8.2

 

10.8

 

Bachelor’s degree

2.1

 

3.3

 

n

6,203

 

7,374

 

Note: All data from 1940–2012 US Census and American Community Survey.

Annual optician earnings are substantially higher (approximately $7,000) on average in states that have optician licensing statutes than in states that do not regulate the profession. Focusing on states adopting new licensing statutes in the sample, we also compared the mean of optician wages before and after licensing was implemented. Among this subset of states, average annual optician earnings were $33,275 before licensing and $35,981 after licensing. Does the strictness of the licensing statute have any discernible economic effects? Table 3 depicts the comparison of states, grouped by the number of exams required.3 The table suggests a positive correlation between the number of exams that prospective opticians must pass to practice and opticians’ annual earnings.
Table 3

Comparison of Annual Optician Earnings ($2012) in the 1940–2012 US Census and American Community Survey by the Number of Exams Required for Licensure

Item

Number of exams required

 

0

1

2

3

4

Mean annual earnings

$29,765

$35,166

$36,143

$37,005

$43,137

Median annual earnings

$26,316

$33,333

$32,000

$33,333

$40,000

n

6,203

342

2,555

4,284

193

Source: Carpenter et al. [2012], and each state’s licensing board and licensing statutes.

Are there other possible explanations for these differences in earnings besides regulations?4 A larger percentage of opticians are males in states that require licensing than in states without licensing. This may partially explain the discrepancy in wages between the two groups. In addition, opticians in states that require licensing have more education than those in states that do not. It is also possible that the states that require licensing have other unobservable differences from the states that do not require licensing. To investigate this possibility further, we focus on states that have adopted licensing legislation, and then compare earnings before and after licensing legislation was adopted. States are grouped by the decade in which they passed an optician law, and wages are relative to states that do not license opticians. As a comparison, we also plot earnings in Texas. A relative wage of 1 would suggest that earnings in the group of states with optician regulations are the same as those in states without optician regulations. One would expect to see relative earnings higher than 1 in states that have passed licensing.

Figure 2 shows an increase in relative earnings either immediately or soon after the passage of licensing legislation. One exception is states that enacted optician licensing between 1970 and 1979 — we could not track this group completely as a result of a lack of observations. The figure shows no evidence of a similar effect in Texas after the passage of certification (in 1976); in fact, optician earnings in Texas fell after certification. If the primary mechanism whereby licensing legislation increases earnings was a higher-quality service, one would expect to see certification having a similar effect on earnings. Of course, this comparison assumes that the majority of Texas opticians obtain certification, and also that consumers can distinguish between different levels of quality in optician services.
Figure 2

Trends in Relative Optician Wages, 1940–2000: (a) states that Licensing Legislation before 1940; (b) States that Enacted Licensing Legislation between 1940 and 1949; (c) States that Enacted Licensing Legislation between 1950 and 1959; (d) States that Enacted Licensing Legislation between 1970 and 1979; (e) States that Enacted Licensing Legislation between 1990 and 1989; (f) texas (Enacted Certification Legislation in 1976)

Source: Licensing data are from Carpenter et al. [2012], and each state’s licensing board and licensing statutes. All other data are from the 1940–2000 US Census.

Note: Relative wages are wages in the selected states divided by wages in states that do not license opticians.Data are not available for states that enacted licensing legislation during the 1960s or for years 1940, 1950, and 1970 for states that enacted licensing legislation during the 1970s.

The Texas Opticians’ Registry provides data on the current number of certified opticians. Comparing this with the total number of opticians currently practicing in Texas shows that only 107 of the 3,761 opticians practicing in Texas are certified (roughly 2.8 percent). Thus, one can presume that opticians in Texas do not feel that certification is worthwhile — and this suggests that consumers may not be able to distinguish between differing levels of quality in optician services.

Correlation of optician licensing and quality of service

Measuring the quality of service delivered to consumers is difficult. Quality can be subjective and difficult to measure precisely. The fact that opticians in Texas decide to forgo obtaining certification suggests that consumers may not be able to distinguish differences in quality levels of services provided by practitioners. Is there any additional evidence that this is the case? Previous studies [e.g., Kleiner [2006]; Kleiner et al. [2014]] have used malpractice insurance premiums as a proxy for the quality of health-care services delivered to consumers. This measure is not perfect — malpractice insurance premiums have been found to be correlated with the amount of procedures performed by health professionals [Baicker et al. 2007] — a symptom of defensive medicine. If licensing enhanced quality, we would suspect that malpractice insurers would charge lower rates. We gathered data on malpractice insurance premiums in 2014 from Lockton Affinity Health. We obtained quotes for each state for employed opticians with 3 or more years of experience working 40 hours per week. The quotes for licensed and unlicensed states were exactly the same (Virginia’s rate was $25 more per year, but this was the only exception); malpractice insurance providers did not appear to consider a lack of licensing a risk factor. This result is very similar to the results of both Kleiner [2006] and Kleiner et al. [2014]. Admittedly, malpractice insurance premiums are not a perfect proxy for quality and could vary for reasons unrelated to the quality of optician services delivered. Our rough examination of quality does not indicate that consumers or malpractice insurance providers can distinguish between the quality levels of services delivered by licensed and unlicensed opticians. We cannot fully ascertain whether this inability results from a lack of sufficient information about the quality of service or simply the fact that the quality of service provided by licensed and unlicensed opticians is essentially the same.

REGRESSION RESULTS

To reach a better understanding of the economic effects of optician licensure, we estimate earnings regressions of the following type:

where earnings are reported by individual i living in state s at time t. I represents a vector of individual control variables (such as age, gender, race, Hispanic origin, and education). Each of these variables is known by economists to be correlated with earnings [Mincer 1958]. FE represents a vector of time and state fixed effects. Over such a long time period, it is likely that the core assumption of DID estimation — namely that all potential state-specific unobserved confounders are time invariant — is likely to be violated. We therefore also include real personal income per capita in each state at time t as an additional control variable to avoid capturing any spurious increase in earnings that results from state economic factors not related to licensing. In addition, we estimate regressions utilizing state specific linear and quadratic trends.5

Our main variable of interest, L, represents a vector of variables used to measure the effects of licensing. We measure licensing by the number of exams required to obtain a license and by the number of days of education and experience required (coding those states without licensing as 0). We also use a simple dummy variable equal to 1 if optician licensing was in effect at time t, 0 otherwise. Finally, we measure the effects of licensing by using the number of years that the statute has been in effect (a variable that we call licensing duration). It is possible that the effects of licensing may take some time to be realized or that the effects of licensing may simply change over time. For example, many licensing statutes include grandfather provisions for opticians already practicing when a new licensing statute is passed. If a large number of practitioners can forgo the licensing process, the effects of licensing may be significantly delayed.

We report results of our estimations using two-way (state and time) fixed effects (with the personal income per capita control), as well as linear and quadratic state-specific trends. Table 4 contains the results of the estimation of the effects of optician licensing. In columns 1 and 2, we use a license dummy variable. The more rigorous specification including linear and quadratic trends indicates that optician licensing is associated with a 12 percent increase in optician earnings. The effect gets slightly larger with the more rigorous specification — suggesting that unobserved time varying state variables are biasing the effect of licensing downwards. We also use license duration (or the number of years that the licensing statute has been in effect) as the measure of licensing. When controlling for two-way fixed effects only, there is evidence of licensing increasing optician earnings by 0.5 percent. This effect becomes statistically insignificant after including linear and quadratic linear trends. Accounting for duration may not be important when measuring the effects of optician licensing — the effects may materialize quickly in this particular labor market. With regard to specific requirements of the licensing statute, in Columns 5–8 we estimate the effects of licensing on optician earnings on the basis of the number of exams and the number of days of education and experience required (in hundreds of days).6 When just controlling for two-way fixed effects, there is evidence that stricter licensing increases optician earnings. Each additional 100 days of experience and education required increases earnings by approximately 2 percent. After controlling for linear and quadratic state trends, there is also evidence that an increase in the number of exams required to obtain licensure is also associated with higher (5.4 percent) optician earnings.
Table 4

Estimates of the Effects of Optician Licensing on the Natural Logarithm of Optician Earnings ($2012)

 

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

Licensing Dummy

0.0369

0.121**

      
 

(0.0369)

(0.0482)

      

Licensing duration

  

0.00515***

−0.000721

    
   

(0.00140)

(0.0139)

    

Number of exams required

    

0.0266

0.0543**

  
     

(0.0186)

(0.0226)

  

Days of education required (100s)

      

0.0224***

−0.00637

       

(0.00904)

(0.0177)

Individual controls

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

State and time fixed effects

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Linear and Quadratic State Trend

No

Yes

No

Yes

No

Yes

No

Yes

R 2

0.2

0.21

0.2

0.21

0.2

0.21

0.2

0.21

n

13,577

13,577

13,577

13,577

13,577

13,577

13,577

13,577

Source: All licensing data are from Carpenter et al. [2012], and each state’s licensing board and licensing statutes. All other data are from the 1940–2000 US Census and 2001–2012 American Community Survey.

Note: Standard errors are adjusted for state clustering. Observations from the state of Texas (which has a certification law) are excluded from the analysis. Individual controls include real personal income per capita, age, and dummy variables for race, gender, Hispanic origin, and education.

*Significance at the 10 percent level; **Significance at the 5 percent level; ***Significant at the 1 percent level.

To further explore the effects of licensing on earnings, we also performed estimations using a set of dummy variables to measure licensing duration — licensure in effect for 1–5 years, licensure in effect for 6–10 years, and licensure in effect for more than 10 years. We also constructed an index of the strictness of licensing — we simply added the number of exams required and the number of days (in hundreds of days) of education and experience required. Table 5 reports the results of these estimations. In Columns 1 and 2 of Table 5, we see consistent evidence that licensing is associated with higher optician earnings within the first 5 years after licensing takes effect. Our estimates range from 11.5–16.9 percent, and the estimated effect increases after controlling for linear and quadratic state trends. This estimate is similar in magnitude to previous nationwide estimates of the effects of licensing [e.g., Kleiner and Krueger 2010], and estimates for single occupations that are also not licensed in every state [e.g., Thornton and Timmons 2013]. There is also evidence that licensing is associated with higher earnings in states where licensing has been in effect for more than 10 years, but the effect is slightly smaller (12.6 percent). In general, the effects of optician licensing appear to materialize very soon after licensing takes effect. This could be an indication that the initial “shock” of licensing discourages entry, but as prospective opticians familiarize themselves with the licensing process (i.e., how to study for the licensing exam) that the effects of licensing fade over time. Another possibility is that increased competition from online retailers has eroded the earnings benefits that opticians previously enjoyed. This result also might shed some light on the question of whether licensing enhances quality. If optician licensing is indeed increasing quality (which we cannot definitively prove or disprove), we would suspect that the effects of licensing would sustain themselves over time — assuming that consumers are willing to pay higher prices for higher quality optician services. Our results suggest that this is not the case in the optician market. In columns 3 and 4, we see some evidence that stricter licensing (either in the form of 100 more days of education and experience requirements or another exam) is associated with an increase in optician earnings (1.7 percent), but this effect is not present after controlling for linear and quadratic state trends.
Table 5

Additional Estimates of the Effects of Optician Licensing on the Natural Logarithm of Optician Earnings ($2012)

 

(1)

(2)

(3)

(4)

Licensure (1–5 years)

0.115***

0.169***

  
 

(0.037)

(0.0462)

  

Licensure (6–10 years)

−0.0136

0.0175

  
 

(0.0391)

(0.0372)

  

Licensure (More than 10 years)

0.0134

0.126*

  
 

(0.0450)

(0.0706)

  

License Index

 

0.0166*

0.00378

 
  

(0.00914)

(0.0121)

 

Individual Controls

Yes

Yes

Yes

Yes

State and Time Fixed Effects

Yes

Yes

Yes

Yes

Linear and Quadratic State Trend

No

Yes

No

Yes

R 2

0.2

0.21

0.20

0.21

n

13,577

13,577

13577

13,577

Source: All licensing data are from Carpenter et al. [2012], and each state’s licensing board and licensing statutes. All other data are from the 1940–2000 US Census and 2001–2012 American Community Survey.

Note: Standard errors are adjusted for state clustering. Observations from the state of Texas (which has a certification law) are excluded from the analysis. Individual controls include real personal income per capita, age, and dummy variables for race, gender, Hispanic origin, and education.

*Significance at the 10 percent level; **Significance at the 5 percent level; ***Significant at the 1 percent level.

Our results seem to suggest that the effects of licensing seem to materialize quickly in the optician labor market. This would seem to indicate that younger opticians may potentially benefit more from licensing than older opticians. To explore this possibility further, we stratified our optician sample into three groups: 25 or younger, 26–45, and 46 and older. We then re-estimated the regression for reach group using a simple licensing dummy. Table 6 contains the results of this estimation. Licensing seems to be associated with the highest optician earnings when focusing our attention on younger opticians (25.6 percent). The estimated effect for opticians between 26 and 45 is similar in magnitude to our estimated average effect (15.8 percent). We find no evidence that licensing is associated with higher earnings for older opticians (46 and older).
Table 6

Estimates of the Effects of Optician Licensing on the Natural Logarithm of Optician Earnings — Differences in Age Groups ($2012)

 

25 or younger

26–45

46 and older

License Indicator

0.256**

0.158**

0.0224

 

(0.118)

(0.0686)

(0.100)

Individual Controls

Yes

Yes

Yes

State and Time Fixed Effects

Yes

Yes

Yes

Linear and Quadratic State Trend

Yes

Yes

Yes

R 2

0.20

0.17

0.17

n

2,447

6,640

4,490

Source: All licensing data are from Carpenter et al. [2012], and each state’s licensing board and licensing statutes. All other data are from the 1940–2000 US Census and 2001–2012 American Community Survey.

Note: Standard errors are adjusted for state clustering. Observations from the state of Texas (which has a certification law) are excluded from the analysis. Individual controls include real personal income per capita, age, and dummy variables for race, gender, Hispanic origin, and education.

*Significance at the 10 percent level; **Significance at the 5 percent level; ***Significant at the 1 percent level.

CONCLUSION

This paper presents the estimated effects of optician licensing on optician earnings. We find evidence that licensing is associated with higher optician earnings (as much as 16.9 percent). The size of this effect is consistent with those found in the existing literature for professionals lacking autonomy in employment. The effects of licensing in the optician labor market seem to be strongest in the first 5 years that licensing takes effect, and among younger opticians. In the only state with certification, Texas, we find no evidence of a similar increase in earnings. In our rough examination of quality, we are unable to find evidence that optician licensing has improved the quality of services delivered to consumers. Taken together, the results indicate that optician licensing is increasing the earnings of professionals with little evidence of benefit accruing to consumers. States with optician licensing laws may wish to reconsider the costs and benefits associated with stringent regulation of the profession.

Footnotes

  1. 1.

    We also investigated the effects of optician licensing on the earnings of optometrists over the same time period (1940–2012) and found no evidence that optician licensing has affected the earnings of optometrists. Ophthalmologists are not identifiable in the Census or ACS — all medical doctors are included in the category of “physicians.”

  2. 2.

    Imputed hourly wages were examined through self-reported hours worked and weeks worked. Some of the data on weeks worked were gathered in intervals, so the medium of the range was used. The results do not substantially change if wages are used as opposed to annual earnings. Also, any changes in the sample (for instance, restricting the analysis to the period 1940–2000) made no material difference in the results.

  3. 3.

    Ideally we would also control for the difficulty of each exam, but data on exam difficulty from 1940 to 2012 is not available.

  4. 4.

    We also estimated the effects of optician licensing on employment population ratios in each state. We found little evidence of any effect. We believe that optician licensing requirements discourage entry into the profession, but the ACS and Census do not provide sufficient data to explore this hypothesis further.

  5. 5.

    Each regression also includes person weights (representing the number of persons represented by each observation in the sample) provided by the Minnesota Population Center’s Integrated Public Use Microdata Series (http://www.ipums.org).

  6. 6.

    The procedure described in the Appendix of License to Work was used to transform education requirements into days. In licensing statutes, education requirements are often reported as “hours” or “clock hours.” For the conversion, hours were divided by 30 (reflecting a 6-hour school day and assuming 5 days of school per week) to convert education requirements into weeks. Weeks were then converted into days by multiplying by 7. If licensing requirements are expressed as years, the number is multiplied by 365. If expressed in terms of degrees, the standard completion time of two years is assumed for an associate’s degree and then multiplied by 365.

References

  1. Adams, Frank, John Jackson, and Robert Ekelund. 2002. Occupational Licensing in a “Competitive” Labor Market: The Case of Cosmetology. Journal of Labor Research, 23(2): 261–278.CrossRefGoogle Scholar
  2. Arrow, Kenneth. 1971. Essays in the Theory of Risk-Bearing. Chicago: Markham Publishing.Google Scholar
  3. Baicker, Katherine, Elliott S. Fisher, and Amitabh. Chandra. 2007. Malpractice Liability Costs and The Practice of Medicine in the Medicare Program. Health Affairs, 26(3): 841–852.CrossRefGoogle Scholar
  4. Benham, Lee, and Alexandra Benham. 1975. Regulating Through the Professions: A Perspective on Information Control. Journal of Law and Economics, 18(2): 421–447.CrossRefGoogle Scholar
  5. Carpenter II Dick M. et al 2012. License to Work: A National Study of Burdens from Occupational Licensing (Arlington, VA: Institute for Justice, May), http://www.ij.org/LicenseToWork.
  6. Cox, Carolyn, and Susan Foster. 1988. “The Costs and Benefits of Occupational Licensing,” Bureau of Economics, Federal Trade Commission, http://www.ftc.gov/system/files/documents/reports/costs-benefits-occupational-regulation/cox_foster_-_occupational_licensing.pdf (accessed February 1, 2015).
  7. Federman, Maya, David Harrington, and Kathy Krynski. 2006. The Impact of State Licensing Regulations on Low-Skilled Immigrants: The Case of Vietnamese Manicurists. American Economic Review, Papers and Proceedings, 96(2): 237–241.CrossRefGoogle Scholar
  8. Feldman, Roger, and James Begun. 1978. The Effects of Advertising Lessons from Optometry. Journal of Human Resources, 13: 247–262.CrossRefGoogle Scholar
  9. Freidman, Milton. 1962. Capitalism and Freedom. Chicago: University of Chicago Press.Google Scholar
  10. Gellhorn, Walter. 1976. The Abuse of Occupational Licensing. University of Chicago Law Review, 44(1): 6–27.CrossRefGoogle Scholar
  11. Haas-Wilson, Deborah, and Elizabeth Savoca. 1990. Quality and Provider Choice: A Multinomial Logit-Least-Squares Model with Selectivity. Health Services Research, 24(6): 791–809.Google Scholar
  12. Han, Suyoun, and Morris Kleiner. 2015. “Analyzing the Duration of Occupational Licensing on the Labor Market”. Unpublished working paper, University of Minnesota.Google Scholar
  13. Kleiner, Morris M. 2000. Occupational Licensing. Journal of Economic Perspectives, 14(4): 189–202.CrossRefGoogle Scholar
  14. Kleiner, Morris. 2006. Licensing Occupations: Ensuring Quality or Restricting Competition. Kalamazoo, MI: Upjohn Institute.CrossRefGoogle Scholar
  15. Kleiner, Morris. 2013. Stages of Occupational Regulation: Analysis of Case Studies. Kalamazoo, MI: Upjohn Institute.CrossRefGoogle Scholar
  16. Kleiner, Morris, and Alan Krueger. 2010. The Prevalence and Effects of Occupational Licensing. British Journal of Industrial Relations, 48(4): 676–687.CrossRefGoogle Scholar
  17. Kleiner, Morris, and Alan Krueger. 2013. Analyzing the Extent and Influence of Occupational Licensing on the Labor Market. Journal of Labor Economics, 31(2): S173–S202.CrossRefGoogle Scholar
  18. Kleiner, Morris, Allison Marier, Kyoung Won Park, and Coady Wing. 2014. “Relaxing Occupational Licensing Requirements: Analyzing Wages and Prices for a Medical Service.” NBER Working Paper 19906.Google Scholar
  19. Leland, Hayne. 1979. Quacks, Lemons, and Licensing: A Theory of Minimum Quality Standards. Journal of Political Economy, 87(6): 1328–1346.CrossRefGoogle Scholar
  20. Maurizi, Alex, Ruth Moore, and Lawrence Shepard. 1981. Competing for Professional Control: Professional Mix in the Eyeglasses Industry. Journal of Law and Economics, 24(2): 351–364.CrossRefGoogle Scholar
  21. Maurizi, Alex. 1974. Occupational Licensing and the Public Interest. Journal of Political Economy, 82(2): 399–413.CrossRefGoogle Scholar
  22. Mincer, Jacob. 1958. Investment in Human Capital and Personal Income Distribution. Journal of Political Economy, 66(4): 281–302.CrossRefGoogle Scholar
  23. Optician Training. 2014. “Optician Job Description, Top Requirements that Should Always be Included in the Optician Job Description,” http://www.opticiantraining.org/optician-job-description/ (accessed June 9, 2014).
  24. Rottenberg, Simon. 1962. The Economics of Occupational Licensing, in Aspects of Labor Economics, edited by Harold Groves Princeton, NJ: Princeton University Press, 3–20.Google Scholar
  25. Ruggles, Steven et al 2010. Integrated Public Use Microdata Series: Version 5.0. [Machine-readable database] Minneapolis, MN: University of Minnesota.Google Scholar
  26. Shapiro, Carl. 1986. Investment, Moral Hazard, and Occupational Licensing. Review of Economics Studies, 53(5): 843–862.CrossRefGoogle Scholar
  27. Smith, Adam. (1776) 1994. Wealth of Nations. New York: Modern Library Edition.Google Scholar
  28. Tedeschi, B o d. 2004. Conflict over Contact Lenses New York Times, June 21.Google Scholar
  29. Thornton, Robert, and Edward J. Timmons. 2013. Licensing One of the World’s Oldest Professions: Massage. Journal of Law and Economics, 56(2): 371–388.CrossRefGoogle Scholar
  30. Timmons, Edward, and Robert J. Thornton. 2010. The Licensing of Barbers in the USA. British Journal of Industrial Relations, 48(4): 740–757.CrossRefGoogle Scholar
  31. Timmons, Edwards, and Robert Thornton. 2008. The Effects of Licensing on the Wages of Radiologic Technologists. Journal of Labor Research, 29(4): 333–346.CrossRefGoogle Scholar
  32. The White House. 2015. “Occupational Licensing: A Framework for Policymakers.” July, https://www.whitehouse.gov/sites/default/files/docs/licensing_report_final_nonembargo.pdf (accessed August 1, 2015).

Copyright information

© EEA 2016

Authors and Affiliations

  1. 1.Department of Business Administration, Saint Francis UniversityLorettoUSA
  2. 2.Mercatus Center at George Mason UniversityArlingtonUSA

Personalised recommendations