Skip to main content
SpringerLink
Log in

Is there a copper super-cycle?

  • Original Paper
  • Published:
Mineral Economics Aims and scope Submit manuscript

Abstract

It has been argued that the copper price exhibits “super-cycles” with periodicities of 30–60 years. Estimates of these super-cycles are typically generated by the bandpass filter. These filtered results are better interpreted as irregular episodic waves. The cyclical interpretation relies on an arbitrary separation between the cycle and an implicitly defined smooth trend. Analysis using the unobserved components model shows the price trend to be stochastic, not smooth. There is some evidence for short and poorly defined cycles but at best weak evidence for a long copper super-cycle. This finding is supported by examination of the price spectrum and generalizes to other non-ferrous metals.

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

Access this article

Log in via an institution

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

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12

Similar content being viewed by others

Data availability

The data used in this paper are deposited at https://smbiena.wixsite.com/website.

Notes

  1. “Are we witnessing the start of a new commodities supercycle?”, Institutional Investor, 22 March 2021. “Are we about to enter a commodity supercycle?”, Forbes, 13 April 2021. “Markets weigh prospects of a new commodities supercycle. Prices from corn to copper are rallying as economies reopen but sceptics think it will be transient”. Financial Times, 12 May 2021. “Is a commodities supercycle under way? Some analysts think a years-long boom has begun, but prices have wobbled of late”, The Economist, 2 June 2021.

  2. https://www.lexico.com/definition/cycle

  3. https://www.lexico.com/definition/wave

  4. Quoting Alker (1981).

  5. The CF bandpass routine replaces the original series x1, …, xT by their deviations from a linear deterministic trend. The bandpass trend estimates add back this component.

  6. See Sect. 4. Cuddington and Jerrett (2008) deflated by the US CPI while I deflate by the US PPI. I find peaks in 1851, 1910, and 1969 and troughs in 1891, 1934, and 1998. These are in line with the turning points in Fig. 1 of CJ (2008).

  7. The commodities literature emphasizes that price processes should be expected to be nonlinear—see Williams and Wright (1991) and Deaton and Laroque (1992). Pérez-Alonso and Di Sanzo (2011) analyze nonlinear (threshold) responses in simple UCM models. The focus of these discussions is on relatively short-term responses. The focus of the current paper is on long cycles which limits the relevance of those contributions.

  8. Cuddington and Jerrett (2008) discuss the choice of deflator in an appendix. They note that the choice of deflator depends on the purpose of the exercise elect to follow Heap (2005) and deflate by the US CPI. I prefer to deflate by the PPI since this eliminates the impact of changes in indirect taxation. Deflation by the PPI supports interpretation in terms of the price copper relative to all wholesale prices. See Buchholz et al. (2019) for a similar argument.

  9. The LME Spot Settlement price or the front NYMEX-Comex price. Subsequently, the Shanghai Futures Exchange (SHFE) also became important. These prices are closely arbitraged (Jin et al., 2021). A large proportion of forward contracts specify delivery up to 12 months forward with the price determined by the average exchange price in the month of delivery or the immediately preceding month, plus typically small regional and contract-specific premiums (Gilbert, 2021).

  10. The price drift is set to -½% per annum and the variances \({\upsigma }_{\upvarepsilon }^{2}\),\({\upsigma }_{\upeta }^{2}\), and \({\upsigma }_{\mathrm{v}}^{2}\) in Eqs. (2–5) are defined by the estimates reported in columns 2 and 3 of Table 11.

  11. The simulations were performed in Gauss and employed the random number function rndn.

  12. The algorithm is modified to allow for turning points prior to the first and subsequent to the final zero year.

  13. I also have data on the price of aluminum in New York from 1895. Aluminum prices fell sharply in the decades prior to the First World War as the consequence of the 1886 invention and subsequent dissemination of the production Hall-Heroult process. Even subsequently, they exhibit a much weaker correlation with copper and the other three base metal prices. The results for aluminum over the 100 years 1921–2020 (not reported) are similar to those for the base metals.

  14. “It should be noted that in a super cycle, prices rise on a trend basis. There are still business cycles within a super cycle” (Heap, 2005, p.5).

  15. Similarly, one could apply the bandpass filter to the residuals from a regression of the copper price, for example, on an index of commodity prices or of the value of the US dollar.

  16. 2011–13, Comex-NYMEX front contract plus Midwest Premium (MWP): Metal Bulletin-Fastmarkets and Platts, (MMACP16), respectively.

    2014–18, MWTP: Platts (MMCUT16).

    2019–20, Comex-NYMEX front contract plus Midwest Premium (MWP): Metal Bulletin-Fastmarkets.

  17. Source: World Bureau of Metal Statistics, World Metal Statistics, and Metal Bulletin-Fastmarkets.

  18. International Monetary Fund, International Financial Statistics.

  19. US Bureau of the Census (1989), series E52. This is the Warren and Pearson (1933) series.

  20. US Bureau of the Census (1989), series 23, updated to 2020 using IMF, International Financial Statistics.

  21. 1800–2016, Federal Reserve Board of St. Louis, https://fred.stlouisfed.org/series/wppiuka, updated using IMF, International Financial Statistics.

Abbreviations

AIC:

Akaike Information Criterion

BIC:

Bayesian Information Criterion

CPI:

Consumer Price Index

GBP:

British pound

IMF:

International Monetary Fund

LME:

London Metal Exchange

ML:

Maximum likelihood

MWTP:

Midwest transaction price

NYMEX:

New York Mercantile Exchange

PPI:

Producer Price Index

PPP:

Purchasing Power Parity

SHFE:

Shanghai Futures Exchange

UCM:

Unobserved Components Model

USD:

US dollar

USGS:

US Geological Survey

WW1:

First World War

WW2:

Second World War

References

Download references

Acknowledgements

This paper originated in a (virtual) lecture at the Vietnam National University, Hanoi, 17 July 2021. I am grateful to Isabel Figuerola-Ferretti, Stephen Pollock, two anonymous referees and the editor of this journal for comments on the initial draft.

Author information

Authors and Affiliations

  1. Johns Hopkins University SAIS Europe, Bologna, Italy

    Christopher L. Gilbert

Authors
  1. Christopher L. Gilbert
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Christopher L. Gilbert.

Ethics declarations

Conflict of interest

The author is a member of the editorial board of Mineral Economics. He is also a non-executive director of the CRU Group Ltd. CRU was not involved in this research and the views expressed do not necessarily represent those of CRU.

Additional information

Publisher's note

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

Appendices

Appendix

Table 11 Unobserved components model estimates (copper)
Full size table
Table 12 Base metals UCM results 1885–2020 (local level model)
Full size table

Data appendix

New York copper prices

Data for 1850–2010 from Edelstein (2011, page 42), who references the US Geological Service (USGS) for prices up to 1896 but trade journals thereafter. Cuddington and Jerrett (2008) attributed their price data to Heap (2005) who referenced the USGS. I update Edelstein’s figures to 2020 by the Midwest Transaction Price (MWTP) which has become the major reference price for commercial copper transactions in the United States.Footnote 16 The prices given by Edelstein are equal to the MWTP in the years immediately prior to 2010.

New York lead, nickel and zinc prices

Data for 1840–2000 from USGS, https://pubs.usgs.gov/sir/2012/5188/tables/

Lead prices updated from 2000 from Platts (Lead Producer North America, MMAAL16) to 2014 and from Metal Bulletin Fastmarkets (LME Settlement plus Midwest Premium) for 2015–20.

Nickel prices updated from 2000 from Platts (Nickel Cathode delivered US, MMAAQ16) to 2018 and for Metal Bulletin Fastmarkets (LME Settlement plus Midwest Premium) for 2019–20.

Zinc prices updated from 2000 from Platts (Zinc SHG delivered US Midwest, MMABD16) to 2018 and for Metal Bulletin Fastmarkets (LME Settlement plus Midwest Premium) for 2019–20.

London copper prices

Cordero and Tarring (1960, page 467) list annual average GBP LME prices from 1877–1956. The LME was closed over the Second World War with trading in copper only resuming in 1953 (Cordero and Tarring, 1960, pages 198–224). Cordero and Tarring (1960) report administered prices over this period. They also report average London prices from 1870–76, prior to the opening of the LME, which they nevertheless describe as LME prices, and high and low prices from 1800–69 (Cordero and Tarring, 1960, page 463). I approximate annual averages over this period by averaging the high and low prices. I update from 1957 to 2020 using annual averages of the LME Settlement Price.Footnote 17 LME copper contracts moves from a GBP to a USD basis in July 1993. USD prices were converted back to GBP at the spot exchange rate.Footnote 18

Deflators

United States: Wholesale Price Index (1800–1890)Footnote 19 and Producer Price Index (from 1891).Footnote 20

UK: Producer Price Index.Footnote 21

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Gilbert, C.L. Is there a copper super-cycle?. Miner Econ (2022). https://doi.org/10.1007/s13563-022-00355-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s13563-022-00355-x

Keywords

Search

Navigation