On September 8, 2026, Japan's Cabinet Office published its second preliminary estimate of gross domestic product for April–June 2026. With it came a short commentary, in Japanese. Its first sentence gives four numbers: real 0.4%, and in brackets, annual rate 1.4%; nominal 1.3%, and in brackets, annual rate 5.5%.
We are going to read that release. This is not investment advice, and we take no view on whether any of this is good or bad. We are translating a government document.
The four headline numbers
All four numbers are seasonally adjusted, and all four compare April–June with January–March. None of them is a comparison with a year earlier. The first figure in each pair is the change from the previous quarter; the figure in brackets is that same change expressed as an annual rate.
- Real GDP, from the previous quarter: 0.4% (first preliminary estimate: 0.3%)
- Real GDP, at an annual rate: 1.4% (first preliminary estimate: 1.1%)
- Nominal GDP, from the previous quarter: 1.3% (first preliminary estimate: 1.2%)
- Nominal GDP, at an annual rate: 5.5% (first preliminary estimate: 4.8%)
The commentary says both the real and the nominal growth rate were revised up from the first preliminary estimate, which was published on August 17, 2026.
How does 0.4 become 1.4? A note in the data tables gives the formula: 年率表示の成長率＝｛(当期の実数÷前期の実数)の4乗－１｝×１００. Divide this quarter's level by the previous quarter's level, raise the result to the 4th power, subtract 1, and multiply by 100.
The levels are printed too, in units of 1 billion yen (10億円), at an annual rate.
- Real GDP, in chained 2020 prices: about ¥598.95 trillion (printed as 598,950.5)
- Nominal GDP: about ¥689.22 trillion (printed as 689,219.1)
The conversion to trillions and the rounding are ours.
The contribution table
A contribution is the part of the growth rate that is attributed to one item. It is measured in percentage points. Everything in this section is real, and measured from the previous quarter. The table first splits the 0.4% in two.
- Domestic demand: −0.1 percentage points (first estimate: −0.2)
- Net exports of goods and services: 0.5 percentage points (first estimate: 0.5)
Below that, each row has two figures: the item's own change from the previous quarter, and its contribution. The English labels are the document's own.
- Private consumption: 0.0%; contribution 0.0
- Private residential investment: −0.6%; contribution printed as −0.0
- Private non-residential investment: −0.9%; contribution −0.2
- Change in private inventories: no growth rate is given for this row; contribution 0.3
- Government consumption: 1.7%; contribution 0.3
- Public investment: −0.5%; contribution printed as −0.0
- Change in public inventories: no growth rate; contribution −0.5
- Exports of goods and services: 0.4%; contribution 0.1
- Imports of goods and services: −1.7%; contribution 0.4
A note in the data tables explains the signs on the import row. Imports are a deduction item in GDP, so their contribution is shown with the reverse sign. In the note's own example, when imports increase, the contribution carries a minus sign. Another note says that, because of rounding, the contributions do not necessarily add up to the growth rate.
Prices: the GDP deflator
Real and nominal differ by prices. The release measures them with the GDP deflator.
- GDP deflator, from the previous quarter, seasonally adjusted: 1.0% (first estimate: 0.9%)
- GDP deflator, from the same quarter a year earlier: 2.6% (the same as in the first estimate)
- Domestic demand deflator, from the previous quarter: 1.3% (first estimate: 1.2%)
The release does not print the gap between the nominal and the real growth rate, and we do not compute one.
The reasons the commentary gives
The commentary gives reasons for revisions. Private non-residential investment was revised from −1.2% to −0.9%. The commentary says this was due, among other things, to the demand-side estimates used as placeholders in the first estimate being revised up, once a survey of corporations for April–June, and other data, were reflected. It does not give a reason for any of the movements themselves.
What the release does not say
Three things this release does not tell you, and so we will not either.
- It does not say why consumption, investment or inventories moved as they did.
- It does not say whether these numbers are high or low.
- It gives no forecast.
What it does print is a schedule. The first preliminary estimate for July–September is due on November 16, 2026. The release adds that the schedule may change.
What it does say is this. In April–June 2026, by the Cabinet Office's second preliminary estimate, Japan's real GDP was up 0.4% on the previous quarter, or 1.4% at an annual rate. Nominal GDP was up 1.3%, or 5.5% at an annual rate.
The Cabinet Office publishes two estimates for every quarter. We will read it again.
Source
Cabinet Office, Economic and Social Research Institute — 2026年4～6月期四半期別ＧＤＰ速報（２次速報値）, summary of results, September 8, 2026: https://www.esri.cao.go.jp/jp/sna/data/data_list/sokuhou/files/2026/qe262_2/pdf/gaiyou2622.pdf. Commentary (ポイント解説), same date: https://www.esri.cao.go.jp/jp/sna/data/data_list/sokuhou/files/2026/qe262_2/pdf/qepoint2622.pdf. Data tables (CSV): https://www.esri.cao.go.jp/jp/sna/data/data_list/sokuhou/files/2026/qe262_2/gdemenuja.html
Growth rates are seasonally adjusted changes from the previous quarter unless labelled as an annual rate or as a change from a year earlier. Contributions are in percentage points. The levels in trillions of yen are LeakSquare's conversion from the printed units of 1 billion yen. This is a reading of a public document. It is not investment advice and makes no recommendation to buy or sell any security.