Skip to content
Descriptive case study · 4-minute read

Tax expense and cash tax — why do the numbers differ?

A case study of three companies illustrates the difference between tax expense and actual payments, and the effect of weighting choices.

MacroVision editorial team · Updated:

In one minute

In the three-company 2017 sample, tax expense/pretax profit is 20.86% and cash tax/pretax profit is 11.79%. Small-sample summaries, not estimates for Korea's economy.

Tax expense/profit before tax20.86
Cash tax/profit before tax11.79
Percentages; 2017 only, three companies, separate financial statements. Each bar is a ratio of sums. Samsung accounts for 92.08% of the sample’s pretax profit.

What is tested?

Nine company-year observations from original separate Samsung, Hyundai and Dong Wha statements, reconstructed without processed company-dataset tax metrics. Respective windows: 2016–2018, 2017–2019 and 2015–2017.

Two numbers, two meanings

The accounting tax ratio divides tax expense by pretax profit; the cash tax ratio divides income-tax payments by pretax profit. Accounting recognition differs from payment. Timing and other reconciling items may explain the gaps; this study did not establish company-specific causes.

Who determines the average?

2017 is the only common year. Tax sums/profit sums give 20.86% accounting tax and 11.79% cash tax; equal-company averages give 22.06% and 16.15%. Samsung generates 92.08% of sample pretax profit and dominates ratios of sums.

Even a three-year average involves a choice

For Hyundai in 2017–2019, the average annual cash tax ratio is 43.28%, compared with 27.97% for total payments divided by total profits. The first gives each year equal weight; the second weights years by pretax profit. Choose the method to fit the question and disclose that choice.

What can we not conclude?

The companies were selected for feasibility, not representativeness. Separate financial statements, different periods and cash taxes relating to other periods prevent conclusions about tax-reform effects, public revenue or the ability to fund benefits in a world with less work.

Sources and transparency

This publication explains the existing research in accessible language. Its accounting example and scenarios are neither official forecasts nor investment recommendations.

Back to the public-funding question ←

What do you think?

What should accompany a tax rate: cash flow, deferred tax or each company's weight?

Further reading

Data guide · AI and work

Which occupations are most exposed to AI? The full ILO ranking

Office and clerical work at the top of the list, physical work at the bottom. What the International Labour Organization's index really measures, and what it does not say about layoffs.

Empirical finding · AI and work

Is AI already eliminating jobs? What the data from Europe show

We examined 13 countries and 39 occupational groups: did exposed occupations lose more jobs in places where firms adopted AI faster? The answer for now: no effect can be detected.

Research finding · Simulation

Even when benefits protect them, those who lost their jobs fall behind

In all 42 matched comparisons in the automation simulation, displaced workers ended below the path in which the change did not happen, even when benefits maintained their level of consumption.

Research scenario

If AI replaces jobs, what will we live on?

An unemployment rate of 20% raises a question beyond the number of jobs: how are the gains from automation distributed?

Accounting illustration

Prices have fallen. Will the benefit buy more?

Purchasing power also depends on wages, taxes and debt. A simple example shows why lower prices are not the whole story.

Descriptive findings

Can corporate tax fund a world with less work?

The available data help frame the question, but do not yet provide a causal answer or guarantee tax revenue.

Scenario analysis

20% global unemployment: an inevitable crisis?

An extreme scenario can lead to very different outcomes — depending on income distribution, debt and public financing.

Economic explainer

Cheaper machines, cheaper products?

The link between automation and prices depends on competition, non-labor costs and how quickly the economy adjusts.

Concept guide

Could a basic benefit preserve living standards?

A purchasing power test must include prices, debts, taxation and the source of the benefit's financing.

Distributional analysis

Asset ownership versus debt: who is more resilient to automation?

The difference between labor income and capital income may become decisive — but asset owners are also exposed to risk.

Behind the research

Behind the forecast: how do you build a model for the age of automation?

From scenario assumptions through government accounting to sensitivity tests — and what still cannot be concluded.

Research finding · Simulation

Falling prices do not reach everyone: why renters are hurt more

In the automation simulation, when savings pass through to prices, renters and mortgage holders fall behind. Indexing the benefit to housing costs helps only when there is budget.

Follow the research

You can also follow new publications via an RSS feed, without leaving any details.

Open publication feed ←

Basic site measurement counts page requests and reaching the article end after reading time. No measurement cookies, personal identifier or IP address in measurement records. Counts are not unique readers.