This is our 7th blog post for our Job Market Paper Series blog for 2025-2026.
Rubina Hundal is a PhD candidate at the University of Chicago Harris School of Public Policy. Her research is in applied microeconomics, with a focus on labor markets and development, studying how information frictions and signals shape access to opportunity, and how targeted interventions can relax those constraints.
We spend years talking about education, training, and skills as the engines of mobility. But what if something smaller and more visible, like your smile, quietly decides whether you even get a chance?
In India’s entry‑level market, oral health is visible, and class coded. Missing or discolored teeth are common among low‑income youth and instantly legible in a headshot. Employers rarely think of themselves as judging by appearance, but they are moving fast with little verified information. My job market paper asks a simple question that sits right at this first screen: how much does visible oral appearance change who moves forward, and through what channel?
The first gate, and why the headshot norm is the right testbed
The roles I study are the ones young people encounter commonly: retail stores, front desk, hospitality, delivery person, call centers. Hiring is quick. Walk‑ins, staffing agencies, and WhatsApp groups dominate. At the first pass, the file is a short resume and a headshot. For customer‑facing positions a smiling photo is often requested. In many walk‑ins the HR assistant even snaps a quick phone photo for the manager. That institutional fact is my testing ground. Because a headshot is routinely visible at the gate, oral appearance becomes part of how candidates are screened before any skill is measured. The paper is therefore about two things. First, it measures the first‑glance penalty. Second, it uses the headshot convention itself to test that penalty by changing only what the photo shows while holding identity and text constant.
A natural suggestion is to not smile at all. In this market, that backfires. Customer-facing jobs expect visible warmth, and recruiters interpret a tight-lipped or neutral expression as distant or lacking confidence. In my perception experiment, the no-smile version of the same face performs nearly as poorly as the poor-teeth version on the employability index. A closed-mouth smile does not fix the problem because the cue being inferred is friendliness. The relevant comparison at the first gate is therefore smiling with healthy teeth versus smiling with visibly poor teeth.
Experiment 1: What people infer at a glance
To isolate visual signals cleanly, I use synthetic AI-generated headshots where a single feature changes at a time (Figure 1). In a nationwide experiment, each respondent saw six faces drawn from the same identities with one randomized edit: no smile, smile, oral condition, darker skin, weight, or glasses. I then asked for ratings that matter in hiring: confidence, diligence, trustworthiness, intelligence, education; as well as those that shouldn’t matter in hiring: wealth, health, attractiveness.
What moves when only oral appearance changes? Almost everything. Faces with visibly poor teeth score about 0.20 standard deviations lower on an employability index built from trust, confidence, and diligence. Every underlying trait shifts in the same direction. Respondents also predict roughly 9 percent lower monthly wages for the exact same face when the teeth look poor. Not smiling carries a similar penalty (Figure 2). Effects for darker skin or obesity are smaller and mixed with the obese male variant performing well, but not the obese female variant. The message is that oral appearance is salient and carries heavy interpretive weight in precisely the traits employers say they value at this stage.
Right: Employability index by headshot variant: Bad teeth and no smile score the lowest.
Experiment 2: do those impressions change actual hiring?
Perceptions are interesting only if they translate into decisions. So, I ran a large, incentivized resume‑rating study with more than 800 hiring professionals from small and medium‑sized firms. Each recruiter evaluated a set of entry‑level profiles. Six profiles came with photos and two to three were text‑only. Photo conditions were randomized within the set: healthy teeth, poor teeth, darker skin, or obesity. Resume fields randomized degree, a short skill certificate, and marital status. Recruiters scored each profile on a 0 to 10 hire‑interest scale. They also made paid forecasts about how other employers, customers, and coworkers would respond to the same candidate in customer‑facing, back‑office, and remote roles. These incentives reward accurate beliefs and keep attention high.
The result is striking. The same resume with visibly poor teeth loses about 0.165 points on the 0 to 10 hire scale relative to healthy teeth. This penalty is about half the size of removing an additional skill certificate from the resume. At common cutoffs that is a 6 to 7 percentage point drop in the probability of advancing. It is largest for customer‑facing jobs and close to zero in back‑office or remote roles. Thus, teeth matter in hiring not only because of hiring managers themselves, but because they price in how customers and coworkers will react. Because recruiters reported forecast beliefs under incentives, I can ask how much of the penalty reflects their own productivity judgments versus concerns about others. When I include their forecasts of customer acceptance and coworker comfort, about half of the bad‑teeth penalty on hiring is absorbed. That pattern is consistent with what I call externality‑mediated screening. Employers internalize how customers will react and how a team will feel, and that gets priced into first‑stage evaluations. Two diagnostic facts help interpret the mechanism. First, review times are flat across photo conditions, roughly one minute per profile. This is not about extra deliberation. It is a fast, first‑glance penalty. Second, the entire hire‑score distribution shifts left for poor‑teeth photos (Figure 3). It is not a small number of extreme downgrades but a broad reweighting against the candidate.
Profiles with poor‑teeth photos in red are shifted left of those with good teeth in blue, showing a broad decline in hiring interest.
From skill gaps to teeth gaps
We often explain inequality in early jobs through differences in education, experience, or networks. What I find here points to another piece of that story, one that’s visible but easy to miss. When oral health becomes a marker of class, even qualified candidates can be screened out before they have a chance to show what they know. The penalty is strongest for young workers applying to customer-facing roles, where employers expect warmth and confidence to be part of the job. In that sense, a confident smile becomes a small advantage that the poor quite literally cannot afford.
What to do about it
Unlike race or gender, oral appearance is a trait workers could change, which makes discrimination itself an equilibrium outcome. If a smile can influence hiring, then basic oral care is not just a health issue, it matters for economic mobility. The problem is less about employers using photos and more about workers entering the market with preventable, visible dental problems that signal disadvantage.
In India, dental visits for the are rare for low-income youth. In fact, in my conversations with nonprofits in this area, I find that even regular brushing and basic oral care is not common in most households. School check-ups are limited, and people usually go to unqualified quackers when there is pain, long after damage becomes visible. Skilling and placement programs could easily add a short oral-health module or link trainees to affordable clinics. Cleanings and simple fillings cost very little compared to what a first job is worth. Even small changes in information and access could help level how candidates present themselves when they first apply. These are modest, practical steps that do not ask firms to change how they hire. They simply make sure that a young person’s chances depend more on skill than on whether they could afford to fix a tooth.
My complementary descriptive evidence from a small, nationally diverse survey of Indian adults suggests that many people underestimate the importance of oral appearance for jobs and cite cost and fear of treatment as primary barriers to dental care. This gap between perceived and actual returns helps explain why visible oral health problems persist even when the labor-market penalty is large. When visible traits are fixable but costly, inequality persists not just because of discrimination, but because of incorrect beliefs about returns.
Rethinking what opportunity looks like
I show that when appearance reflects unequal access to care, it quietly reinforces inequality that already exists. I use the headshot norm to study this process at the exact point where looks first enter the hiring decision. The takeaway is simple: development policy often focuses on schools and training, but sometimes opportunity also depends on smaller, everyday barriers like information about the importance of oral health or even the cost of a dentist visit. Reducing those barriers will not just improve health; it can change who gets seen as employable in the first place.
Featured image: Generated using Google Gemini
