Skip to content

OECD data via eolas

The OECD publishes standardised macroeconomic indicators for its 38 member economies, harmonised so they're directly comparable across countries. eolas serves five NZ indicators from the OECD Key Economic Indicators (KEI) dataflow — chosen because they're the headline series most NZ analysts cross-reference internationally.

For deeper OECD coverage (other countries, non-KEI dataflows), see the external links at the bottom.

What's in the catalogue

Dataset Frequency Description
nz_cpi Quarterly Consumer Price Index, base 100. OECD-harmonised — directly comparable to other OECD CPIs.
nz_gdp_growth Quarterly Real GDP growth, percent change quarter-on-quarter.
nz_unemployment Quarterly Harmonised unemployment rate (15-64), percent of labour force.
nz_interest_short Monthly Short-term interest rate, typically the 3-month inter-bank rate.
nz_interest_long Monthly Long-term interest rate, typically 10-year government bond yield.

These are all part of the OECD's KEI dataflow (OECD.SDD.STES,DSD_KEI@DF_KEI). The same dataflow has equivalents for every OECD member — so any analysis using these five datasets can be replicated for other countries by changing the country code in the underlying SDMX query.

Refresh schedule

Weekly, Wednesday morning NZ time. The OECD itself updates KEI on its own monthly / quarterly cadence; our weekly refresh catches new releases within seven days of publication.

meta = client.info("nz_cpi")
meta["last_refreshed_at"]
meta["source_last_modified_at"]   # OECD's own publish timestamp where available

License

OECD data is not CC-BY. The OECD Terms and Conditions permit non-commercial use and academic / research / journalism use with attribution, but prohibit commercial redistribution without a separate licence agreement.

What this means in practice:

  • Free analysis, charts, blog posts: fine, attribute OECD.
  • Charging customers for OECD data: not fine without an OECD agreement.
  • eolas's stance: we serve OECD data on the Free and Pro tiers as a convenience for analytical use. Enterprise customers are excluded from automated OECD redistribution — if you're building a paid product on top of OECD data, contact us so we can structure the licence appropriately, or query OECD's own API directly.

Recommended attribution: "Source: OECD, via eolas (eolas.fyi). © OECD."

Common patterns

NZ CPI vs OECD average

The whole point of using OECD data is cross-country comparability. Here, NZ CPI alone:

import matplotlib.pyplot as plt

cpi = client.oecd("nz_cpi", start="2010-01-01")
cpi.plot(x="date", y="value", title="NZ CPI (OECD harmonised)", figsize=(10, 4))
plt.show()

library(ggplot2)

cpi <- eolas_get_oecd("nz_cpi", start = "2010-01-01")
ggplot(cpi, aes(date, value)) +
  geom_line() +
  labs(title = "NZ CPI (OECD harmonised)", y = "Index, 2015=100")

For multi-country comparisons (NZ vs Australia vs US), you currently need to query the OECD KEI dataflow directly — that's on the eolas roadmap as a separate client.oecd_compare() helper.

All four interest-rate / GDP / unemployment series at once

indicators = ["nz_cpi", "nz_gdp_growth", "nz_unemployment", "nz_interest_short"]
dfs = {name: client.oecd(name, start="2010-01-01") for name in indicators}

import matplotlib.pyplot as plt
fig, axes = plt.subplots(2, 2, figsize=(12, 8))
for ax, (name, df) in zip(axes.flatten(), dfs.items()):
    df.plot(x="date", y="value", ax=ax, title=name, legend=False)
plt.tight_layout()
plt.show()

library(patchwork)

indicators <- c("nz_cpi", "nz_gdp_growth", "nz_unemployment", "nz_interest_short")
plots <- lapply(indicators, function(name) {
  df <- eolas_get_oecd(name, start = "2010-01-01")
  ggplot(df, aes(date, value)) + geom_line() + labs(title = name)
})
Reduce(`+`, plots) + plot_layout(ncol = 2)

Pipeline use

OECD datasets are full-snapshot — the upstream SDMX source replaces the whole table on each revision. Each of the five KEI datasets is under 500 KB; re-downloads are fast.

Note: OECD data may not be used for commercial redistribution without an OECD agreement (see License above). Use sync_bulk / eolas_sync_bulk for analytical pipelines; do not feed OECD data into a paid product without checking your licence position.

result = client.sync_bulk("nz_gdp_growth", path="/data/nz_gdp_growth.parquet")
print(result.status)  # "downloaded" (first run) or "unchanged"

result <- eolas_sync_bulk("nz_gdp_growth", path = "/data/nz_gdp_growth.parquet")
result$status  # "downloaded" or "unchanged"

eolas sync nz_gdp_growth --out /data/nz_gdp_growth.parquet

See the Bulk downloads guide for scheduling and format options.

Source-specific notes

  • Vintage / revision: OECD revises its data with new vintages — what was Q1 2024 CPI yesterday may be slightly different tomorrow as Stats NZ submits revised inputs. We pull the latest revision each refresh; if you need a frozen-in-time snapshot, archive your local copy.
  • Country code: every OECD series is internally keyed by an ISO country code (NZL for New Zealand). The five datasets we expose are pre-filtered to NZ; the country code is in the metadata but not in column names.
  • OECD's KEI vs other OECD products: KEI is the harmonised, comparable, monthly/quarterly indicator set. OECD also publishes annual data (national accounts, social indicators), structural metrics (labour, education), and policy indicators — none of those are currently in eolas. Roadmap if there's demand.
  • No Stats NZ-OECD divergence within reason: OECD's NZ CPI mostly tracks Stats NZ's published CPI, but small differences appear due to OECD harmonisation rules (e.g. owner-occupied housing treatment). Use OECD when comparing to other countries; use Stats NZ for the NZ-domestic narrative.

Where to find more