Skip to main content

Advertisement

Log in

ITER: A Quarterly Indicator of Regional Economic Activity in Italy

  • Research Paper - Italy and Europe
  • Published:
Italian Economic Journal Aims and scope Submit manuscript

Abstract

This work documents the construction of the new quarterly indicator of regional economic activity (Indicatore Trimestrale dell’Economia Regionale—ITER), which uses a parsimonious set of regional variables and combines them by means of temporal disaggregation techniques to obtain a quarterly index that is consistent with the official data on national and regional GDP and marked by a small lag compared with the reference period. The methodology was implemented to produce quarterly indicators for the economies of Italy’s four macro-areas in the period 1995–2017. With a view to assessing the performance of the quarterly indicator, a forecasting exercise was conducted regarding annual GDP growth in the four macro-areas for the period 2014–17. The forecasting performance of ITER is in line with that of the indicators developed by other national research institutions.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2

Similar content being viewed by others

Notes

  1. In Italian the ITER acronym stands for "Indicatore Trimestrale dell'Economia Regionale".

  2. Before May 2018, flash estimates of Italy’s GDP were released after 45 days.

  3. According to the Istat Census of 2011 the geographical distribution of Italian industrial districts was as follows: 45, 37, 38 and 21 districts in North-East, North-West, Centre and South areas, respectively.

  4. In this exercise we consider as target variable the first release of the annual regional GDP.

  5. The estimations were carried out by means of the Fernández and SSC functions included in the Quilis (2013) Matlab Temporal Disaggregation Toolbox. At present, the Proietti model has not been implemented, as the corresponding Matlab function in the Toolbox only covers the case of a single short-term indicator.

  6. We follow the approach developed in Petrella and Piselli (2012).

References

  • Bacharach M (1965) Estimating non-negative matrices from marginal data. Int Econ Rev 6(3):294–310

    Article  Google Scholar 

  • Barbone L, Bodo G, Visco I (1981) Costi e profitti in senso stretto: un’analisi su serie trimestrali, 1970–1980. Bollettino della Banca d’Italia, n 36:465–510

    Google Scholar 

  • Chow G, Lin AL (1971) Best linear unbiased distribution and extrapolation of economic time series by related series. Rev Econ Stat 53(4):372–375

    Article  Google Scholar 

  • Cuevas A, Quilis M, Espasa A (2015) Quarterly regional GDP flash estimates by means of benchmarking and chain linking. J Off Stat 31:627–647

    Article  Google Scholar 

  • Di Fonzo T (1990) The estimation of M disaggregate time series when contemporaneous and temporal aggregates are known. Rev Econ Stat 72(1):178–182

    Article  Google Scholar 

  • Fernández RB (1981) Methodological note on the estimation of time series. Rev Econ Stat 63:471–478

    Article  Google Scholar 

  • Pappalardo C, Piselli P (2002) Stima ed analisi del ciclo economico territoriale 1987–2000. Rivista di politica economica 92(5/6):201–246

    Google Scholar 

  • Petrella A, Piselli P (2012) Ricostruzione delle serie regionali di produzione industriale. mimeo, Banca d'Italia

  • Proietti T (2002) La stima rapida dei conti economici territoriali. In: atti della VI Conferenza Nazionale di Statistica (Rome, November 2002)

  • Proietti T (2006) Temporal disaggregation by state space methods: dynamic regression methods revisited. Econ J 9:357–372

    Google Scholar 

  • Quilis EM (2013) A Matlab library of temporal disaggregation and interpolation methods. Mimeo, Macroeconomic Research Department, Ministry of Economy and Competitiveness, Madrid

  • Rossi N (1982) A note on the estimation of disaggregate time series when the aggregate is known. Rev Econ Stat 64:695–696

    Article  Google Scholar 

  • Santos-Silva JMC, Cardoso F (2001) The Chow-Lin method using dynamic models. Econ Model 18:269–280

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Valter Di Giacinto.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

Appendix 1

1.1 Dataset and seasonal adjustment

In the following we briefly describe the time series considered in our empirical application, the source and the treatment for the seasonality.

Number of employed workers (quarterly series; availability: about two and half months delay. Sources: Istat and INPS). The total number of both independent and payroll employees resident in the region; this series is also available for the main sectors. The measurement of payroll workers is net of the full-time equivalent number of workers receiving benefits under the Cassa integrazione guadagni (wage supplementation) program.

Export of goods (quarterly series; availability: about 3 months delay. Source: Istat). Volume of exports; this series is also available for the main sectors.

Electricity consumption (monthly series; availability: about 1 month delay. Source: Terna). Total amount of electric energy dispatched quarterly (in GWh).

Regional Index of Industrial Production (monthly series; availability: about 1 month delay. Source: our calculations based on Istat data). See Appendix 2 for a detailed description.

Business demographics (monthly series; availability: about 1 month delay. Source: our calculations based on Infocamere data). The number of new firms recorded in the public registry in each quarter, net of transfers; the nominal component may be reconstructed for the main sectors.

Bank loans to the private sector (monthly series; availability: about 1 month delay. Source: Banca d’Italia). End-of-quarter outstanding amounts of bank loans to firms and households, deflated by the consumer price index; the nominal component may be reconstructed for the main sectors.

Residential transactions (quarterly series; availability: about 2 months delay. Source: Agenzia delle Entrate—Osservatorio sul mercato immobiliare—Tax revenue Agency). The number of residential transactions concluded in each quarter in the real estate market.

Foreign Tourism (monthly series; availability: about 1 month delay. Source: Banca d’Italia). Quarterly expenditure of foreign tourists.

Google Trends (monthly series; availability: about 1 month delay. Source: Google). Google measures searches with the keyword ‘disoccupazione’ (unemployment) for people dwelling within the region in each quarter.

The series were seasonally adjusted, using TRAMO-SEATS, when needed according to pre-testing analysis.

Appendix 2

2.1 The Regional Index of Industrial ProductionFootnote 6

In this Appendix we present the methodology that we use to construct the regional series for industrial production, which hinges on a procedure of extrapolation and re-aggregation of national series for the industrial production at the sectoral level. The seasonally adjusted national series are available from 1990 onwards on a monthly basis for the 15 sectors of the ATECO 2007 two-digit classification of economic activities. To approximate the contribution of the industrial sectors of each region to the corresponding national index, we use data on sectoral value added at current prices taken from the territorial economic accounts. To begin with, we calculate the industrial production index for sector s in region R (i.e. \(IP_{s}^{R}\)) as:

$$ IP_{s}^{R} = \frac{{VA_{s}^{R} }}{{\sum\nolimits_{R} {VA_{s}^{R} } }}IP_{s}^{ITA} , $$

where \(VA_{s}^{R}\) denotes the regional sectoral value added and \(IP_{s}^{ITA}\) the Italian industrial production aggregate. The regional index is subsequently obtained by re-aggregating the individual sectoral regional indices, weighed by the share of each sector on the total industrial value added of each region, that is:

$$ IP^{R} = \sum\nolimits_{s} {\frac{{VA_{s}^{R} }}{{\sum\nolimits_{s} {VA_{s}^{R} } }}IP_{s}^{R} } . $$

The \(IP_{{}}^{R}\) index may be deemed able to capture the common cyclical component underlying regional dynamics in manufacturing activity, obtained by properly weighting the national sectoral trends.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Di Giacinto, V., Monteforte, L., Filippone, A. et al. ITER: A Quarterly Indicator of Regional Economic Activity in Italy. Ital Econ J 7, 129–147 (2021). https://doi.org/10.1007/s40797-020-00131-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s40797-020-00131-2

Keywords

JEL Classification

Navigation