oneapi
Sign in
Financiar v1 / Convergenta

Interest-rate convergence criterion

The reference value from Protocol No 13 — the average long-term rate of the three best-performing member states in terms of price stability, plus 2 percentage points — and where Romania stands against it. Both quantities are read as twelve-month averages.

The selection is made on INFLATION, not on interest rates, and that is the trap this route exists for. The lowest long-term rates in December 2025 belonged to Denmark, Sweden and Germany; the three best performers on prices were Cyprus, France and Italy. A computation done on the lowest rates would give a lower reference value and a harsher verdict than the real one.

The route excludes nothing. The European institutions drop from the selection states considered outliers — Greece in 2013, in deflation — and that is a judgement, not a formula. Here the computation is mechanical, the states chosen are named, and the full ranking comes alongside, so that anyone wanting a different selection can make it. The result is not the official verdict of the Commission or of the ECB.

How much the exclusion matters: in June 2013, the mechanical computation picks Greece — inflation 0.23%, but a long-term rate of 15.31% at the height of the crisis — and gives a reference value of 8.83%, which Romania passes at 6.15%. The institutions excluded Greece; without it the reference is 5.24%, and Romania does not pass. The verdict flips. That is why exclude exists: the choice stays with the client.

The aggregates (EA, EU27_2020) do not enter the selection — they are averages, not states — but they do appear in the ranking, flagged, because they are useful reference points.

The inflation criterion is not computed here: it too requires a reference value built the same way, but its verdict depends on the same outlier exclusions, and about those we have nothing published to decide on.

GET
https://oneapi.ro/api/financiar/v1/convergenta/criteriu-dobanda

Query parameters

NameTypeDescriptionAccepted values
lunaoptional string | null Optional: the month, YYYY-MM, from 1997 onwards. Without it, the last computable month — the end given in meta.acoperire. e.g. 2013-06e.g. 2024-06
excludelistoptional text Optional: entities dropped from the SELECTION, not from the ranking. Comma-separated, at most 10 (?exclude=EL). e.g. EL

How it behaves

  • The reference value from Protocol No 13 to the Treaty — the average long-term rate of the three best-performing Member States in terms of price stability, plus 2 percentage points — and where Romania stands against it. Both quantities are read as twelve-month averages.
  • The selection is made on INFLATION, not on interest rates, and that is the trap this route exists to close. In December 2025 the lowest long-term rates belonged to Denmark, Sweden and Germany; the three best on prices were Cyprus, France and Italy. A calculation done on the lowest rates would give a smaller reference value and a harsher verdict than the real one.
  • The route excludes nothing on its own, and the result is NOT the official verdict. The European institutions drop from the selection any state judged an outlier — a judgement, not a formula.
  • How much that matters: in June 2013 the mechanical calculation picks Greece — inflation 0.23%, but a long-term rate of 15.31% at the height of the crisis — and yields a reference value of 8.83%, which Romania clears at 6.15%. The institutions excluded Greece; without it the reference is 5.24% and Romania does not clear it. The verdict flips. Hence exclude: the choice stays with the client, and clasament carries every entity so it can be made knowingly.
  • Inflation is the ratio of the average index over the last 12 months to the average over the 12 before — the protocol's definition — not the month-on-same-month-last-year change, which jumps with every price shock. The long-term rate is the average of the monthly observations over the same window.
  • Aggregates are not states and cannot be "the best three". EA and EU27_2020 are averages; they stay in the ranking, flagged este_stat_membru: false, because they are useful benchmarks.
  • Greece is EL at Eurostat and GR at the ECB. Codes are kept as each source writes them, and cod_bce bridges them. exclude takes the code from entitate; exclusi_necunoscuti shows which requested code matched nothing, so a reflexive GR does not pass for a successful exclusion.
  • The computable period ends where the first of the two series ends — usually the HICP index, published later than the rates. meta.acoperire says so.
  • The inflation criterion is not computed here: its reference value is built the same way, but its verdict depends on the same outlier exclusions, and there is nothing published to decide those from.

What it is for

  • Where Romania stands against the interest-rate criterion for euro adoption.
  • Rebuilding the calculation with a different selection of reference states.
  • Comparing EU states' funding costs on a consistent basis.

Response structure

FieldTypeDescription
data.perioada string The month the calculation is made for, YYYY-MM.
data.fereastra object The 12 months the averages are taken over.
data.fereastra.de_la string Start of the covered interval.
data.fereastra.pana_la string End of the covered interval, inclusive.
data.fereastra.luni integer Number of months covered.
data.valoare_referinta string The average long-term rate of the three chosen states, plus 2 percentage points.
data.exclusi array The entities the client excluded, as written.
data.exclusi_necunoscuti array Those among them that match no entity in the ranking — the sign of a wrong code.
data.componente.media_celor_trei string The average long-term rate of the three.
data.componente.marja string The 2 percentage points from the protocol.
data.state_de_referinta[] array The three chosen states, with their inflation and long-term rate.
data.state_de_referinta[].entitate string
data.state_de_referinta[].inflatie string
data.state_de_referinta[].rata_lunga string
data.romania.rata_lunga string The 12-month average of the Romanian long-term rate.
data.romania.inflatie string Romanian inflation over the same window.
data.romania.indeplineste boolean true if the rate is below the reference value. Not an official verdict.
data.romania.distanta string Rate minus reference, in percentage points. Negative means below.
data.clasament[] array Every entity, in ascending order of inflation.
data.clasament[].entitate string
data.clasament[].cod_bce null The ECB code, when it differs from the Eurostat one. null when they match.
data.clasament[].este_stat_membru boolean false for aggregates — they cannot be chosen.
data.clasament[].inflatie string
data.clasament[].rata_lunga string null when the entity has no published long-term rate; it then cannot enter the selection.
meta.surse.inflatie string
meta.surse.rate string
meta.temei string The legal basis for the values or for the way they are computed.
meta.acoperire.de_la string The first period we have data for.
meta.acoperire.pana_la string The last period we have data for.
meta.entitati integer
meta.nota_selectie string A note on how to read the response: a caveat, a convention or a limit of the source.
meta.nota_neoficial string A note on how to read the response: a caveat, a convention or a limit of the source.
meta.nota_excludere string A note on how to read the response: a caveat, a convention or a limit of the source.
meta.nota_medii string How to read the averages in the response and when the median says something else.
meta.nota_agregate string A note on how to read the response: a caveat, a convention or a limit of the source.
meta.nota_grecia string A note on how to read the response: a caveat, a convention or a limit of the source.
meta.fara_date data is null and this says why. The month falls outside the computable period.
meta.fara_date_cod The short, STABLE word for the reason, for clients that branch on it in code: fara_date is prose and may be rephrased, this is not.

Response examples

GEThttps://oneapi.ro/api/financiar/v1/convergenta/criteriu-dobanda
{
    "data": {
        "perioada": "2026-08",
        "fereastra": {
            "de_la": "2025-09",
            "pana_la": "2026-08",
            "luni": 12
        },
        "valoare_referinta": "5.04",
        "exclusi": [],
        "exclusi_necunoscuti": [],
        "componente": {
            "media_celor_trei": "3.04",
            "marja": "2.00"
        },
        "state_de_referinta": [
            {
                "entitate": "DK",
                "inflatie": "1.55",
                "rata_lunga": "2.75"
            }
        ],
        "romania": {
            "rata_lunga": "6.47",
            "inflatie": "8.58",
            "indeplineste": false,
            "distanta": "1.43"
        },
        "clasament": [
            {
                "entitate": "DK",
                "cod_bce": null,
                "este_stat_membru": true,
                "inflatie": "1.55",
                "rata_lunga": "2.75"
            }
        ]
    },
    "meta": {
        "surse": {
            "inflatie": "Eurostat — HICP, indicele total (prc_hicp_minr, ECOICOP ver. 2, TOTAL, 2015 = 100).",
            "rate": "BCE — rata dobânzii pe termen lung pentru convergenţă."
        },
        "temei": "Protocolul nr. 13 la Tratatul privind funcţionarea Uniunii Europene.",
        "acoperire": {
            "de_la": "1997-12",
            "pana_la": "2026-08"
        },
        "entitati": 29,
        "nota_selectie": "Cele trei state se aleg după INFLAŢIE, nu după dobândă. În decembrie 2025 cele mai mici rate lungi erau ale Danemarcei, Suediei şi Germaniei, dar cele mai bune trei la preţuri erau Cipru, Franţa şi Italia — o selecţie făcută pe rate ar da o valoare de referinţă mai mică şi un verdict mai aspru decât cel real.",
        "nota_neoficial": "NU e verdictul oficial. Instituţiile europene exclud din selecţie statele considerate atipice — o judecată, nu o formulă. Aici calculul e mecanic şi nu exclude nimic de la sine.",
        "nota_excludere": "Cât de mult contează: în iunie 2013 calculul mecanic alege Grecia (inflaţie 0,23%, dar rată lungă 15,31% în plină criză) şi dă o referinţă de 8,83%, pe care România o trece cu 6,15%. Instituţiile au exclus-o; fără ea referinţa e 5,24% şi România NU o trece — verdictul se inversează. Parametrul `exclude` lasă alegerea clientului: `?exclude=EL`.",
        "nota_medii": "Inflaţia e raportul dintre media indicelui pe ultimele 12 luni şi media pe cele 12 dinainte, nu variaţia lunii faţă de aceeaşi lună a anului trecut. Rata lungă e media observaţiilor lunare din aceeaşi fereastră.",
        "nota_agregate": "`este_stat_membru: false` marchează agregatele (`EA`, `EU27_2020`). Sunt medii, nu state, deci nu pot fi „cele mai bune trei\"; rămân în clasament ca repere.",
        "nota_grecia": "Grecia se scrie `EL` la Eurostat şi `GR` la BCE. Codurile se păstrează aşa cum le dă fiecare sursă; `cod_bce` face legătura acolo unde diferă. La `exclude` se scrie codul din `entitate` — `EL` — iar `exclusi_necunoscuti` arată ce cod cerut n-a nimerit nimic, ca un `GR` scris din reflex să nu treacă drept excludere reuşită."
    }
}

Error codes

CodeSlugWhen it occurs
401neautentificatToken missing, invalid, expired or revoked.
403feature_indisponibilThe current plan does not include the feature the route depends on.
404negasitThe response gives that period.
422validareA month written other than YYYY-MM or before 1997, or a malformed entity code in exclude.
429cota_depasitaThe plan's monthly quota has been exceeded (see Retry-After).

All errors share the same shape. eroare.cod is stable and is the value you should branch on in code — not the text in mesaj, which may change.

{
    "eroare": {
        "cod": "validare",
        "mesaj": "A month written other than YYYY-MM or before 1997, or a malformed entity code in exclude.",
        "detalii": {
            "luna": [
                "The value sent does not satisfy the parameter rule."
            ]
        }
    }
}