A marketing agency hires two account directors within a month. Both have nine years’ experience. Tomás asks for 12% above the advertised figure, cites another offer and gets 8%. Leila, asked for her expectations, says she is “happy with the range”. She is placed at its lower end.
Three years of percentage-based increases later, the gap between them has grown from 9,000 to more than 11,000. Leila’s client retention is the best in the agency. She finds out about the difference when a new finance assistant accidentally sends a spreadsheet to the wrong list. In the meeting that follows, the managing director explains, truthfully, that Tomás “negotiated well”. It is the worst thing he could have said. He has just told his best account director that her pay reflects a ten-minute conversation in which she was polite.
Managers negotiate pay from an unusual position. They are a party to the bargain, with a budget to protect, and they are also responsible for the fairness of the result across the team. Most are trained for neither role.
The psychology of negotiation is well researched, and much of it is about how predictable mental shortcuts move the final number. This guide covers what a manager needs to know: how anchors work, why some people never ask, what a counteroffer does and how to design a process in which pay tracks contribution instead of bargaining nerve.
The first number matters most.
Anchors pull final pay toward them. Do not let previous salary or bravado set the anchor.
Asking is unevenly distributed and unevenly received.
Pay set by negotiation rewards confidence and circumstance.
Decide pay from the role, then explain it.
Make the first offer the fair offer, and audit for gaps that bargaining has already created.
How do anchors shape salary outcomes?
The first number stated in a negotiation acts as an anchor, and final agreements end up closer to it than they otherwise would. The effect was first shown by Amos Tversky and Daniel Kahneman in the 1970s and has been replicated in settings from estate agency to legal judgments.
Estate agents shown different listing prices for the same house produced valuations that moved with the listing price while denying it had influenced them. In negotiation experiments, whichever side made the first offer tended to obtain a better outcome, because the counterpart adjusted insufficiently from the opening figure.
In hiring, the anchor may be the advertised range, the candidate’s stated expectation or their previous salary. Each has consequences.
| Anchor | Effect | Fairness risk |
|---|---|---|
| Candidate’s previous salary | Offer tracks what the last employer paid | Imports old inequities into your pay structure |
| Candidate’s stated expectation | Confident candidates set high anchors | Rewards self-assurance |
| Advertised range | Bounds the discussion for everyone | Low, provided the range is real |
| Employer’s first offer | Sets the centre of the bargain | Low-balling damages trust if discovered |
This is the reasoning behind laws in a growing number of places that bar employers from asking about salary history. An offer anchored on previous pay means that someone underpaid in their last job is underpaid in yours.
People start from the anchor and adjust away from it until the figure seems plausible, then stop. Because a range of figures is plausible, they stop too early. Knowing about the effect weakens it only slightly.
Who asks, and who does not?
A substantial share of people do not negotiate a job offer at all. Who does is not random. Personality plays a part, as do seniority, culture and whether the person believes negotiating is acceptable in that setting.
The best-known research concerns gender. Linda Babcock and colleagues documented, in work popularised in Women Don’t Ask, that men in their samples were considerably more likely than women to negotiate starting salaries. Later studies have complicated the picture. A large field experiment by Andreas Leibbrandt and John List found that when a job advert stated explicitly that pay was negotiable, the gender difference in negotiating disappeared. Some more recent surveys find women now ask about as often as men but are less likely to succeed.
The reluctance, where it exists, is not irrational.
Evaluators in a series of experiments were less willing to work with candidates who had attempted to negotiate their pay, and the penalty was significantly larger for women than for men. Hesitating to ask can be an accurate reading of a real social cost.
Similar hesitancy has been observed among people from cultures where bargaining with an employer is seen as disrespectful, first-generation professionals who were never told the number is movable, and anyone who badly needs the job. The upshot for managers: if pay is set by who pushes, it will reflect confidence and circumstance as much as value.
What should a manager do about it?
Stop relying on the candidate to set the number.
Decide the offer from the role and the evidence. Before the conversation, place the candidate in the range on the basis of skills and experience against written criteria. Offer that figure. If two candidates with the same profile would get different offers because one asked, the process is rewarding the wrong thing.
Make your first offer your fair offer. The traditional approach, start low and leave room to move, punishes exactly the people who trust you. It also sets up the discovery that Leila made.
If you negotiate, say so to everyone. Stating that the figure is open to discussion removes much of the difference in who asks.
Record the reason for any deviation. If you pay more than the criteria suggest, write down why: a scarce skill, a competing offer. If you cannot write a reason you would be willing to show to the rest of the team, do not do it.
Some organisations go further and adopt a no-negotiation policy: the offer is the offer. The online forum Reddit did so for a period under its then chief executive Ellen Pao, on the argument that negotiation disadvantaged women. Such policies can cost an occasional candidate. They also remove a source of inequity and a great deal of managerial time. The link to pay openness is explored in our guide to pay transparency.
How should you respond when an employee negotiates?
As you would to any raise request: listen, investigate and answer with reasons, the sequence set out in how to say no to a raise request. Two additional points apply to negotiation as such.
First, do not penalise the asking. The evidence on backlash suggests managers do this without noticing, by coming to see the person as demanding. An employee who asks is giving you information you would otherwise get in a resignation letter.
Second, separate the person’s argument from their leverage. “I have delivered X and the market pays Y” is an argument about value and should be assessed on its merits. “I’ll leave if you don’t” is leverage. If you concede to leverage what you refused to argument, the whole team learns which one works.
Should you make a counteroffer when someone resigns?
Sometimes, but less often than managers do. A counteroffer is a negotiation conducted at the worst possible moment: the employee has already decided to go, the manager is anchored on the competing offer and on the pain of a vacancy, and both are acting under time pressure.
| Question to ask yourself | If yes | If no |
|---|---|---|
| Would I have paid this before they resigned? | The offer revealed you were behind; correct it | You are paying for the threat, not the person |
| Is money the real reason they looked? | A counter may work | The other reasons will resurface |
| Can I do this without creating an inequity? | Proceed, and review peers | You are trading one resignation for several |
| Do I trust them to stay committed? | Proceed | The relationship has already changed |
Loss aversion drives much counteroffer behaviour. The imminent loss of a known colleague feels bigger than the abstract cost of a pay anomaly, so managers overpay to avoid it. Recruiters often claim that most people who accept counteroffers leave within a year anyway. The precise figures quoted are rarely well sourced, but the logic is sound: money was seldom the only reason, and a raise does not fix the others. Those reasons are examined in why employees really quit.
Every counteroffer is observed. If the only reliable way to get a market adjustment in your team is to produce a competing offer, your best people will go and get one, and some of them will take it.
What about negotiation tactics aimed at you?
Managers face the same psychology from the other side. Candidates and employees, sometimes coached, will anchor high, cite offers and set deadlines. A few habits help.
- Prepare your own number first. An anchor has least effect on someone who has already done the analysis and written the figure down.
- Re-anchor on the range. “Our range for the role tops out at 84,000” replaces their number with a structure.
- Ask what the number is based on. A well-founded figure survives the question. A bluff usually softens.
- Trade, do not just concede. If base pay cannot move, a sign-on payment, a review date or flexibility may address the actual concern. What candidates are weighing is described in job offer psychology.
- Slow down. Time pressure makes people rely more on anchors and first impressions. “Let me come back to you tomorrow” is nearly always available.
How do you fix the gaps negotiation has already created?
By looking, and then paying. An annual review of pay within each level, comparing people with similar roles, experience and performance, will show the gaps that trace back to starting salaries. Close them through targeted adjustments separate from the merit budget, so that fixing an inequity does not use up the money meant to reward the year’s work.
Add a “starting-salary check” to every hire: before the offer goes out, compare it with the current pay of the two or three most similar people on the team. If the new hire would be paid more for the same profile, either the offer is too high or the incumbents are underpaid. Deal with whichever is true before it is discovered.
The agency eventually did what it should have done at the start. It set a range for account directors, placed each of them by written criteria, and found four people, three of them women, below where the criteria put them. Leila’s adjustment was the largest. The managing director told her he was sorry it had taken a mis-sent spreadsheet, and that from now on the agency would decide what the job paid before it met the person.
Frequently asked questions
What is anchoring in salary negotiation?
Anchoring is the tendency for the first figure mentioned to pull the final agreement toward it. In pay discussions the anchor may be a previous salary, an advertised range, the candidate’s stated expectation or the employer’s first offer.
Should employers ask candidates about their current salary?
It is unlawful in a growing number of jurisdictions and poor practice elsewhere. Basing an offer on previous pay carries over any underpayment from the last employer. Set the offer from the role’s range and the candidate’s skills.
Should a manager make a counteroffer to keep an employee?
Only if the employee is genuinely underpaid against your own structure and money is the main reason they looked. Countering purely to avoid a vacancy tends to delay the departure, create inequities and teach others that resigning is how to get a raise.
How can companies make salary negotiation fairer?
Publish ranges, place people in them using written criteria, make the first offer a fair one, tell every candidate whether the figure is negotiable, record the reason for any exception and review pay within each level every year to correct unexplained gaps.
Discover more from Kurums | Business Intelligence
Subscribe to get the latest posts sent to your email.