What a 3% COLA Assumption Actually Means If You’re a Google Employee Planning to Retire on RSUs

Most retirement plans don’t fail because people are too aggressive.

They fail because they’re just reasonable enough to go unquestioned.

A Googler asked me this week whether 3% was a solid COLA projection for retirement planning.

It’s a reasonable assumption.

It’s also doing more work than most people realize.

She had a significant position in Google stock, a clear number in her head for what retirement looked like, and a plan built on the idea that prices would rise about 3% a year for the next 25 years.

When we model that scenario, it all worked.

But that one assumption is quietly carrying the entire plan.

COLA, the cost-of-living adjustment, is the annual increase you build into your retirement income to keep up with rising prices.

Here’s what I mean:

The standard inflation numbers you hear about: CPI and PCE measure what a working-age consumer spends money on.

Gas to commute.

Rent or a mortgage.

Groceries. Clothing.

Those categories get tracked, weighted, and averaged into the headline number that shows up on CNBC.

Right now that number is 3.3% annually, spiked by energy costs tied to the war in Iran. Strip out food and energy and the rate is closer to 2.6%.

That’s the number most retirement calculators use.

It’s also not your number if you’re retiring in your mid-40's to-late fifties and planning to live another 40 to 50 years.

 

Retirees spend differently.

The share of spending that goes toward healthcare rises steadily with age. You have doctor visits, specialist care, medications, supplemental insurance, and possibly long-term care.

Housing costs stay sticky because most retirees aren’t moving frequently.

The Bureau of Labor Statistics actually publishes a separate experimental index for elderly consumers called CPI-E, and it runs 0.2 to 0.5 percentage points above regular CPI every single year without fail. That sounds small. But over 30 years of compounding, your purchasing power can decrease faster than your plan assumes.

3% on the headline. 4 to 5% on what you actually spend in retirement. Those are different plans with different endings.

There’s a second problem specific to Google employees, and it’s the one I see most often in these conversations.

RSUs vest in dollars. Google stock has performed well enough over a long enough period that the accumulated position looks large: upwards of $1 million and more.

When you’re looking at a number that size, the retirement math feels comfortable. Generous, even.

There’s a natural tendency to let that comfort substitute for actual planning.

A COLA assumption gets picked where 3% feels reasonable, a withdrawal rate gets assumed, and the spreadsheet looks fine.

But the harder structural questions don’t get asked because the position is big enough that urgency never quite arrives.

The tension worth naming clearly is this: concentrated stock and retirement income are two separate problems that feel like one because they share a number.

How much you have in Google stock tells you what you’re worth on paper today.

It does not tell you how much reliable, inflation-adjusted, tax-efficient income you can actually generate from it across a 30-year retirement.

One is a balance sheet figure.

The other is an engineering problem.

The questions that actually change your retirement picture aren’t complicated, but they require honest answers.

When do you need this money to start working, and at what level?

If you’re retiring before 65, you’re covering healthcare out of pocket until Medicare kicks in, and healthcare inflation is not simple.  Projections can vary from 5 to 8% annually and in ways that have nothing to do with the Iran war or tariff policy.

That’s a materially different income requirement than someone retiring with full employer coverage through 65.

What does your spending actually look like in retirement?

How much is fixed and non-negotiable versus flexible and adjustable if markets turn?

A 3% COLA assumption stress-tested against a 4% reality looks completely different depending on whether you have a $40,000 annual gap to fill or a $140,000 one.  That 1% delta is not marginal.

The macro context matters too, because it sets the floor your assumptions have to clear.

The Iran war and its effect on energy prices is a real example. 

Gasoline spiked 21% in March alone, the largest monthly increase recorded since 1967.

Tariffs are adding a persistent upward pressure on goods prices that isn’t resolving quickly.

The range for inflation over the next decade is probably 3 to 3.5% in a base case and 4 to 5% if the war extends and tariff policy stays aggressive.

Perhaps 8% or higher is probably too pessimistic for a stress-test assumption, that would require a 1970s-style structural breakdown.  But it has happened. 

Today, the US economy, is an energy exporter and is meaningfully more insulated against than it was then. But 3% applied across every category of retirement spending is almost certainly too optimistic, particularly on the healthcare side where the math just doesn’t cooperate.

Here’s the reframe that for Google employees sitting on a large RSU position: the question isn’t whether 3% is precisely right or slightly wrong.

It’s whether your retirement income structure can absorb being wrong, without forcing decisions you don’t want to make at 72.

That’s not a number problem.

It’s a structural one, how the income is built, how much is fixed and inflation-protected, how much is variable, how much depends on Google continuing to perform, and what the tax drag looks like as you start pulling it all down.

So what happens if this doesn’t get examined?

The position keeps vesting. It keeps growing.

And the plan stays comfortable, at least on the surface.

But the options that exist today- Roth conversions, systematic diversification, spreading tax impact across multiple years, structuring income intentionally- those don’t stay open forever. They narrow quietly over time.

 

Doing nothing feels fine when the stock is working and the retirement number looks big enough.

But tax complexity compounds. And the decisions don’t get easier by waiting.

This is the work.

Helping Google employees think through RSU decisions, retirement income, and the inflation assumptions sitting underneath everything else- not with generic advice, but with a structure built for a real career at Google and a retirement that has to hold up for 40+ years.

If your plan looks roughly right, but you haven’t pressure-tested how it behaves when the assumptions drift, that’s worth looking at sooner rather than later.

If you want to keep thinking through this, I write about these decisions regularly.  Click here to learn more.

And if you’re at the point where this is becoming real, not theoretical, you can schedule a conversation and we’ll walk through it together.

 

--

Disclaimer: This is general information, not individualized tax or investment advice. Outcomes depend on your specific situation- please coordinate with your CPA or advisor.