# The Switch-Premium Score, to Search Now, Hold, or Press Internally

*You can pull your field's current switch premium, net out forfeitures, and score your own situation as search now, hold, or press internally.*

- Canonical URL: https://www.refolk.ai/candidates/guides/switch-premium-score
- Pillar: Reading the market
- Format: Framework
- Published: 2026-10-04
- Last reviewed: 2026-10-04
- Reading time: 17 min

You are weighing whether to launch a job search for a raise or stay where you are, and you want to know whether switching will actually pay in your field. This guide is for anyone who has read that changing jobs buys a 10 to 20 percent raise and wants to know if that number is true for them. It gives you a repeatable score: pull your field's current switch premium, convert a likely offer into a net comparable number, and land on one of three verdicts - search now, hold, or press internally.

The reason this guide exists is that the headline number is a national average, and the national average hides almost everything that matters. The job-switcher premium currently runs strongly positive in resources, mining, and construction, and negative in parts of IT and hospitality. A single number read off a top-ranking page can flip your decision in the wrong direction. The fix is to work from the per-field series and then net it against your own move math.

## What is the switch-premium score

The switch-premium score is a three-verdict judgement you reach by comparing your field's current switcher-minus-stayer wage gap to your own net move math. It answers one repeated question: does switching pay enough, right now, in my field, for me specifically?

The score has two inputs. The first is market: what the public wage-growth series say switchers in your sector are getting over stayers this month. The second is personal: what you would net after subtracting everything you forfeit by leaving. You run both, then read the result against a published threshold band. The output is a decision with a re-check date, not a statistic.

**0.7 pp - Long-run job-switcher premium over stayers, 1997 to 2025**

The switcher advantage averages only about 0.7 percentage points across the full Atlanta Fed series history, which makes the 2022 peak the anomaly.

The score matters because two readers in the same month, with the same salary, can reach opposite verdicts. One is a construction project manager with no unvested equity; the other is a software engineer four years into a six-year vesting cliff. The market gap and the forfeiture both move, and you cannot read either off a blog's headline.

> A single national switch-premium number is useless for a specific decision in a specific field.

## The dimensions that drive the score

Four dimensions decide whether switching pays. Score each one before you project an offer, because any single one can sink an otherwise attractive move.

- **Field gap.** The current switcher-minus-stayer wage growth in your sector. This is the ceiling on what a switch can realistically buy you.
- **Source definition.** Whether you are reading a base-pay or gross-pay series, because the two produce very different gaps in the same month.
- **Forfeiture load.** What you lose by leaving now: unvested equity, unvested 401(k) match, an unpaid bonus, a PTO payout, a coverage gap.
- **Market thickness.** How many employers in your field are bidding, which determines how easily a positive field gap turns into a real offer.

Each dimension proves something, and each can lie. The field gap proves the market is paying switchers more, but it lies when you read a 3-month smoothed average that is hiding a just-started reversal. The forfeiture load proves the real cost of leaving, but it lies if you value unvested equity at grant rather than realizable value. Market thickness proves switching liquidity, but it lies if you compare an industry count to a job-title count, which are different query types.

#### What goes into one switch-premium score

1. **Field gap** - Current switcher-minus-stayer growth in your sector, the realistic ceiling on a switch
2. **Source definition** - Base vs gross pay, which changes the gap by percentage points in the same month
3. **Forfeiture load** - Unvested match, equity, bonus, PTO, and coverage gap subtracted from the offer
4. **Market thickness** - How many employers are bidding for your title, which decides if the gap is realizable

*Score from the outside in - the market gap bounds the offer, your move math nets it down, and thickness tells you how likely you are to realize it.*

## Pull your field's current gap from public data

Start with the two series that publish industry detail: the Atlanta Fed Wage Growth Tracker and ADP Pay Insights. Both are monthly, both are free, and together they bracket your field's real number.

The Atlanta Fed tracker is a median year-over-year percent change computed at the individual level and reported as a 3-month moving average. It breaks out by sector groups: construction and mining; education and health; finance and business services; leisure and hospitality; manufacturing; public administration; and trade and transportation. The stayer and switcher sub-series each run monthly from January 1997, so the gap is computable across roughly 355 observations of history.

ADP Pay Insights tracks more than 14.8 million workers monthly and reports base pay and gross pay growth across worker mobility, sector, employer size, pay quartile, and 56 US metropolitan areas. The key difference: ADP gross pay includes overtime and workforce composition, while the Atlanta Fed uses matched hourly wages. That is why, in the same month, the two disagree by percentage points.

| Source | Date | Switcher | Stayer | Gap (pp) |
|---|---|---|---|---|
| Atlanta Fed (hourly, 3mma) | Aug 2026 | 5.0 | 3.6 | 1.4 |
| Atlanta Fed | Jul 2025 | 4.0 | 4.1 | -0.1 |
| ADP (gross pay) | Sep 2026 | 7.3 | 4.4 | 2.9 |
| ADP (base pay) | Sep 2026 | 4.8 | 3.0 | 1.8 |
| Long-run avg | 1997-2025 | - | - | ~0.7 |

Read the table carefully. In July 2025 the gap went negative, with stayers at 4.1 percent edging out switchers at 4.0 percent - the sixth straight stayer-favoring month, a reversal not seen since 2010. That is the whole point of pulling the current number: the sign itself can flip.

> **Rule:** Match the definition to your pay
>
> Use the Atlanta Fed hourly series if you are hourly or your overtime is stable. Use the ADP base-pay series if you are salaried. Do not score a salaried move off ADP gross-pay numbers, which are inflated by overtime you may not work.

For per-field switcher signs, the strongest public statement is older: resources and mining showed the highest switcher gain at 11.8 percent, while leisure and hospitality was the only sector with negative switcher growth at -1.3 percent. Treat those as range markers, not current readings. They trace to older ADP data, and a current, clean per-industry switcher-minus-stayer table from the Atlanta Fed is downloadable but is not republished per field in press summaries. For a field like IT, the direction - developer pay softening - is supported by Payscale, which projects tech pay increases falling from 4 percent to 3.5 percent, but the precise current IT-negative figure is not established publicly in these sources. Where the number is not published, say so and cross-check the latest sector release yourself rather than inventing one.

## Convert a likely offer into a net comparable number

A gross offer is not a comparable number until you subtract everything you forfeit by leaving. The documented method is consistent across finance sources: sum total compensation, then subtract annualized forfeitures and one-time switching costs, and discount any signing bonus to year one only.

The forfeiture that swings most decisions is the 401(k) match. Multiply the employer-funded portion of your balance by your unvested percentage. If the employer-funded portion is worth 20,000 dollars and you are 60 percent vested, you could forfeit about 8,000 dollars by leaving now. A safe-harbor match is a different animal: it is 100 percent vested immediately and cannot be forfeited, so it never enters this math. And in a documented case, a worker at year five of a six-year graded schedule stood to forfeit 74,000 dollars - enough to erase a double-digit raise.

| Tenure | Vested % | Kept | Forfeited |
|---|---|---|---|
| <2 yr | 0% | $0 | $30,000 |
| 2 yr | 20% | $6,000 | $24,000 |
| 4 yr | 60% | $18,000 | $12,000 |
| 6 yr | 100% | $30,000 | $0 |

This table, a six-year graded schedule on a 30,000 dollar match, shows why timing matters more than the offer headline. Leaving at year four forfeits 12,000 dollars; waiting to year six forfeits nothing. Your position on the vesting curve, not the raise percentage, is often the deciding variable.

Beyond the match, inventory the rest: forfeited unvested equity (discounted to realizable value, since a new grant may be illiquid or underwater), this year's bonus if you leave before payout, a PTO payout, relocation, and COBRA at roughly 500 dollars per week for family coverage during any coverage gap. Then remember the new employer has its own vesting clock. You escape one cliff and start another, so treat the new match as worth zero until you clear its vesting.

**Net comparable worksheet**

```
New base salary:                        +$________
New bonus (expected, year one):         +$________
Signing bonus (year one only):          +$________
New equity (realizable, year one):      +$________
New 401(k) match (treat as $0 if unvested): +$0
---
Current base you give up:               -$________
Forfeited unvested 401(k) match:        -$________
Forfeited unvested equity (realizable): -$________
Unpaid current-year bonus:              -$________
Lost PTO payout:                        -$________
COBRA / coverage gap (weeks x $500):    -$________
Relocation and one-time costs:          -$________
---
Net year-one premium ($):               =$________
Net year-one premium (% of current base): =____%
```

*Fill every line in dollars. Negatives are forfeitures. The bottom line is what you actually net in year one.*

> **Watch out:** A signing bonus is not recurring
>
> Model two years, not one. A signing bonus hits only year one, so a package that looks great in year one can collapse in year two once the bonus and any first-year grant roll off. Score the steady-state number, not the recruiting sweetener.

Writing a resume and cover letter tuned to each posting you chase is where a candidate burns the most time before any of this math pays off. [Refolk](/candidates) builds your resume from your own history, tailors it to every posting, and scores how well you actually fit, so the applications behind a positive score cost you minutes instead of evenings.

## The step-by-step scoring procedure

Run the score in order. The market pull comes first so you are projecting against a real ceiling, then the forfeiture math, then the verdict. Finance sources push the forfeiture math first and negotiation sources lead with the threshold; the order below puts the market number ahead of both so your offer projection is grounded before you net it.

#### Score your situation

1. **Pull your field's current gap** - Open the Atlanta Fed tracker and the latest ADP release and record switcher %, stayer %, and the gap for your sector group. Done means one current gap figure per source for your field.
2. **Check staleness** - Note the release month and that both use a 3-month moving average that can lag a turning point by weeks. Done means you know the read is at most one month old and smoothed.
3. **Translate gap to an expected gross offer** - Apply the current switcher premium to your base as a realistic ceiling, not the 2022 peak. Done means a plausible gross offer range.
4. **Inventory forfeitures** - List unvested equity, this year's bonus if you leave before payout, the unvested 401(k) match, PTO payout rules, and any coverage gap. Done means a dated dollar list.
5. **Compute the net comparable** - New total comp minus annualized forfeitures and one-time switching costs, discounting a signing bonus to year one. Done means a net annual premium in percent.
6. **Score against the threshold** - Compare your net premium to the 10-20% band and the ~5-6% floor. Done means one of three verdicts.
7. **Act and set a re-check** - Search now if the net clears the threshold and the field gap is positive, hold if below the floor, press internally if the gap is thin but you have leverage. Done means a decision with a re-check tied to next month's release.

#### From public data to a verdict

1. **Market pull** - Record your field's switcher and stayer growth from both series
2. **Offer ceiling** - Apply the current gap to your base as a realistic maximum
3. **Net the offer** - Subtract forfeitures and one-time costs, discount the signing bonus
4. **Verdict** - Read the net premium against the 10-20% band and 5-6% floor

*The score runs market-first, then nets the offer down, then reads the result against a fixed threshold.*

## Read the score: search now, hold, or press internally

Your net year-one premium, read against the published threshold band, produces the verdict. The dominant guidance is a 10 to 20 percent boost when changing jobs, with a floor near 5.8 percent; a recruiter benchmark is a minimum of 10 percent above current base to make a move worth serious consideration, with 15 to 25 percent for a genuine step up in scope.

- **Search now** when your net premium clears the 10 to 20 percent band and your field gap is positive. The market is paying switchers, and your move math survives the forfeitures.
- **Hold** when your net premium sits below the roughly 5 to 6 percent floor. The offer does not clear the cost of leaving, usually because forfeitures ate it, or because your field gap is thin or negative.
- **Press internally** when the field gap is thin but you have real offer leverage, or when forfeitures erase a raise that would otherwise clear the band. An internal adjustment avoids resetting your vesting clock and opening a coverage gap.

#### Verdict grid

Horizontal axis runs from Thin or negative field gap to Strong positive field gap. Vertical axis runs from Net premium below floor to Net premium clears the band.

| Quadrant | What it means |
| --- | --- |
| Hold | Market is soft and the math does not clear; stay and bank your vesting |
| Press internally | Field is hot but forfeitures eat the move; use the demand as leverage for an internal raise |
| Hold | Even a decent market cannot overcome your forfeiture load right now; wait out a vesting cliff |
| Search now | Market pays switchers and your net clears the band; launch the search |

*Cross your field's current gap against your net move math to land on one of three actions.*

The verdict is never permanent. Both series are monthly and smoothed, so set a trigger to re-pull after the next release. ADP, for instance, publishes on a fixed schedule - its October report was scheduled for November 4 at 8:15 a.m. ET - so you always know when a fresh reading lands.

## How the score goes wrong

Most bad switch decisions come from a handful of repeatable errors, and each has a cheap check. This is the part worth re-reading before you act, because a false positive here costs you a reset vesting clock and a coverage gap for nothing.

| Failure mode | What it looks like | The check |
|---|---|---|
| Quoting the 10-20% national average | Expecting 15% in a field running negative | Pull your sector's current gap before projecting |
| Reading a stale smoothed series | The 3mma hides a just-started reversal | Confirm the release month; look at the single month too |
| Confusing ADP and Atlanta Fed gaps | Scoring a salaried move off gross-pay numbers | Match definition to your situation: base vs gross |
| Counting the signing bonus as recurring | Year one looks great, year two collapses | Model two years; the bonus hits year one only |
| Ignoring the new vesting clock | You escape one cliff and start another | Treat the new match as $0 until you clear its vesting |
| Treating equity at grant value | Forfeited RSUs feel offset by a new grant | Discount both to expected realizable value |
| Using Zippia's -1.3% / 11.8% as current | Scoring off figures that trace to older data | Cross-reference the latest Atlanta Fed and ADP releases |

The first and the last share a root: both treat a published number as if it described this month and your field. The premium is mean-reverting toward about 0.7 percentage points, which is why the 2021 to 2022 peak near 14 to 16 percent was a one-off tied to record openings. Bank of America had already measured the job-change premium falling to roughly 4 percent median by January 2025, down from about 14 percent in 2022. If your mental anchor is still 2022, every score you run will be too optimistic.

> **Tip:** When the two series disagree, that is information
>
> If ADP gross-pay shows a wide gap but the Atlanta Fed hourly series is flat, the extra money is likely overtime and composition, not base pay you can negotiate into your next offer. Score the base-pay number and treat the gross-pay spread as upside, not the plan.

## Market thickness and how to test it

Thickness is how many employers are bidding for your title, and it decides whether a positive field gap is realizable or merely theoretical. A hot gap does you no good if only three employers hire your specialty in your city. The same title can offer very different switching liquidity across markets.

Refolk's index makes this concrete. In Refolk's index of professional profiles there are 352,540 US software engineers against 43,827 in the UK - an 8.0x ratio. More bodies in the pool is a proxy for more employers bidding, which means a positive switch premium is easier to realize in the deeper market. The hospitality industry shows 1,023,895 US professionals in the same index, a very thick pool, though recall that hospitality is exactly the sector where published switcher growth has run negative - thickness and premium are different things.

| Query | Country | Count |
|---|---|---|
| Title: Software Engineer | United States | 352,540 |
| Title: Software Engineer | United Kingdom | 43,827 |
| Industry: Hospitality | United States | 1,023,895 |

One caution on this table: the title rows and the industry row are different query types, so compare within a type, not across. The US-to-UK software engineer ratio is a fair comparison; the hospitality count is not comparable to either, because it counts an industry, not a job title.

To test thickness for your own field before you commit to a search, look at how many people with your exact profile recently switched into the kind of employer you want. A live search of recent movers tells you whether the switching path you are counting on is well-trodden or rare.

Ask me this: `Senior software engineers in the US who changed employers in the last 12 months and now work at companies with 500+ staff.` - [run the search](https://www.refolk.ai/start?q=Senior%20software%20engineers%20in%20the%20US%20who%20changed%20employers%20in%20the%20last%2012%20months%20and%20now%20work%20at%20companies%20with%20500%2B%20staff.).

*Returns recent US senior-engineer switchers now at large employers, a direct read on how liquid that switching path is right now.*

**8.0x - US to UK software engineers in Refolk's index**

The deeper pool, 352,540 versus 43,827, means more employers bidding and a switch premium that is easier to actually realize.

## Keep the score current

A switch-premium score is only as good as its freshness date, so treat it as a living read rather than a one-time calculation. Both headline series update monthly and both smooth over three months, which means a reading can lag a genuine turning point by weeks. Re-pull on each release and re-score before you make an irreversible move.

#### Before you call the verdict final

- [ ] I pulled my sector's switcher and stayer growth from both the Atlanta Fed tracker and ADP this month
- [ ] I recorded the release month and checked the single-month reading, not only the 3-month average
- [ ] I used the base-pay definition if I am salaried and gross-pay only as upside
- [ ] I projected the offer off the current field gap, not the 2022 peak
- [ ] I inventoried every forfeiture in dated dollars: unvested match, equity, unpaid bonus, PTO, coverage gap
- [ ] I treated the new employer's match as $0 until I clear its vesting clock
- [ ] I modeled two years so the signing bonus is not counted as recurring
- [ ] I confirmed my field's switching path is liquid, not just that the gap is positive
- [ ] I set a re-check tied to next month's release date

What to do next depends on your verdict. If you scored search now, open the applications while the gap holds, because a smoothed series means the window may already be narrowing. If you scored hold, set a calendar trigger for the next release and bank any vesting milestone you are approaching. If you scored press internally, bring the net math - not the national headline - to the conversation, because your employer can check the same public series you did. In all three cases, the discipline is the same: re-pull the number, re-net the offer, and let the score, not the headline, decide.

## Frequently asked questions

### Is it worth switching jobs for a raise right now?

It depends entirely on your field and your forfeitures, not the national average. The job-switcher premium currently ranges from roughly 11.8 percent in resources and mining down to negative in hospitality. Pull your sector's current gap from the Atlanta Fed tracker and ADP, apply it to your base as a ceiling, then subtract your unvested match, equity, and any coverage gap. Only move if the net clears the 10 to 20 percent band or at least the roughly 5 to 6 percent floor.

### How much of a raise should I ask for when changing jobs?

Common guidance is 10 to 20 percent over your previous salary, with a floor near 5.8 percent and 15 to 25 percent for a genuine step up in scope. But treat that as a target, not a guarantee. In a field where the switcher gap is thin or negative, asking for 15 percent can price you out before you start. Calibrate the ask to your sector's current gap, then net out what you forfeit by leaving.

### What is the job switcher versus stayer pay gap?

It is the difference in wage growth between people who changed employers and those who stayed. In August 2026 the Atlanta Fed Wage Growth Tracker showed switchers at 5.0 percent and stayers at 3.6 percent, a 1.4 percentage point gap. Over the full series history since 1997 the premium averages only about 0.7 percentage points, and it briefly went negative in July 2025.

### Why do ADP and Atlanta Fed show different switch premiums?

They measure different things. The Atlanta Fed uses matched hourly wages, while ADP gross pay includes overtime and workforce composition. For the same month, ADP gross-pay showed a 2.9 percentage point gap while the Atlanta Fed showed 1.4. Match the definition to your situation: use base-pay comparisons if your pay is salaried and overtime is not a factor.

### How do I account for a 401(k) match I would forfeit by leaving?

Multiply the employer-funded portion of your balance by your unvested percentage. If the employer-funded portion is worth 20,000 dollars and you are 60 percent vested, you could forfeit about 8,000 dollars by leaving now. A safe-harbor match is 100 percent vested immediately and cannot be forfeited. Also treat the new employer's match as worth zero until you clear its vesting clock.

### When should I press for an internal raise instead of leaving?

Press internally when your field's switcher gap is thin but you hold real offer leverage, or when forfeitures erase an otherwise strong raise. If your net premium falls below the roughly 5 to 6 percent floor after netting out an unvested match or equity, an internal adjustment often beats a move that resets your vesting clock and opens a coverage gap.

---

*From the Refolk guide library. I revise these guides rather than replacing them, so the current version is always at https://www.refolk.ai/candidates/guides/switch-premium-score*