Accounting quality in railway companies during the nineteenth and twentieth centuries: the case of Spanish NORTE and MZA

Abstract

Prior literature studying railway accounting during the nineteenth and twentieth centuries defends the thesis of lack of reliability of accounting figures. This prior research, which mainly studies the cases of the United Kingdom and the United States, offers mixed views on the causes, or simply accepts this thesis without providing conclusive evidence, as is the case of historical research in Spain. We provide novel evidence on the quality of railway accounting and contribute to this prior debate by (1) analysing the accounting for two material accruals: depreciation and prior period adjustments; (2) studying the persistence of earnings and its components, and (3) analysing how accrual accounting affects persistence. These analyses are conducted for the period 1856–1939 for the two major Spanish railway companies (MZA and NORTE). The reported evidence suggests that earnings are highly persistent. However, we show that there are significant differences across firms and that these differences are particularly obvious when analysing the adjustments for prior period earnings. Overall, our evidence does not support the thesis that accounting was underdeveloped, but rather, that managerial accounting choices lowered accounting quality.

Keywords:

• accounting history
• accounting quality
• railway accounting
• earnings persistence
• Spanish railways

Acknowledgements

We gratefully acknowledge the helpful comments and suggestions from the editors (Mark Clatworthy and Edward Lee) and two anonymous reviewers. We are also grateful to Miguel Artola, Juan Baños, Mariló Capelo, Salvador Carmona, Nieves Carrera, Delfina Gomes, and Massimo Sargiacomo, as well as seminar participants at the 40th Annual Congress of the European Accounting Association (Valencia), X Esteban Hernández Esteve Meeting of Accounting History (Las Palmas de Gran Canaria), the 11st International Research Seminar on Accounting History (Sevilla), the 1st International Seminar of Accounting History (Siena), and the XXII Workshop on Accounting and Management Control (Orense). We are thankful to the Spanish Railway Foundation for providing access to MZA and NORTE historical files and annual reports.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Beatriz Santos-Cabalgante http://orcid.org/0000-0001-9450-6659

Beatriz García Osma http://orcid.org/0000-0002-6600-8342

Domi Romero Fúnez http://orcid.org/0000-0002-0583-5475

Notes

1 MZA is the acronym for Compañía de los ferro-carriles de Madrid a Zaragoza y a Alicante. NORTE is the acronym for Compañía de los caminos de hierro del Norte de España.

2 Bryer (1991, p. 449) points out the depreciation accounting of rolling stock was the ‘modal practice’ of railway companies in the early 1840s, but Edwards and Boyns (1994, p. 1177, fn. 10) and Arnold and McCartney (2002, p. 206) affirm that just a minority of companies charged depreciation on the track.

3 The world ‘mania’ is used frequently to refer to ‘the elated phase an inherently unstable alternation between excessive optimism and pessimism’ (Bryer 1991, p. 439, quoted in McCartney and Arnold 2003, p. 822). There were three booms of investments in UK best known as railway-mania: 1824–1825, 1835–1837, and 1845–1847.

4 Garcke and Fells (1893, pp. 95–96), Matheson (1893, p. 5) or Dodd (1895, p. 61) [in Edwards (1986, p. 206)].

5 See also Flesher and Previts (2009) for an excellent review of the literature on US railway companies.

6 Prior research in Spain usually focuses on one of the two leading railways and a single phase of the sector. In relation to MZA, the topics include: the treatment of the staff expenses in 1913–1935 (Martínez Vara 2001, 2004, 2006); the internal accounting regulations in 1857–1908 (Fidalgo and Miranda 2006); measurement criteria of fixed assets (Fidalgo et al., 2015); the coding system for expenses in the workshops (Villacorta Hernández and Martínez Vara 2009); the evolution and changes of the Capital Account and of the Operating Account, and the impact of the organisational structure on the accounting statements from 1856 to 1874 (Santos-Cabalgante et al. 2012a, 2012b, 2014). In relation to NORTE: the accounting treatment of depreciation during the twentieth century (Bentabol et al. 2011); the evolution of the Operating Account from 1900–1925 (Montoya and Guzmán 2011). Few studies compare both companies: e.g. the accounting and statistical functions of the Chief of a railway station (Montoya et al. 2012), and the first railway audits (Villacorta Hernández and Müller 2014).

7 Extent of railway open in 1847 (in miles): US (6565), UK (5000), Germanic States – including Denmark and Holland – (4542), France (1722), Belgium (457), Russia (200), and Italy (170) (Lardner 1850, p. 417).

8 Burghers were people belonging to the middle or upper stratum of the middle social class. They usually had certain financial capital and political influence.

9 Previously, in France, the Rothschild family had established the Compagnie du Chemins de Fer du Nord du France, a continuous reference for MZA (De Los Cobos and Martínez Vara 2009); the Pereire brothers had established the Compagnie du Nord (Comín et al. 1998).

10 Until 1881, all General Managers in NORTE were French and had been working in the Pereire’s railway business (Vidal and Ortúñez 2002, pp. 32, 35).

11 For information about the management style and the proprietorship of NORTE and MZA in Ortúñez (1993, 1997), Vidal Olivares (1996, 1997, 1999), and in Vidal Olivares and Ortúñez (2002).

12 In Spain, railway companies were exempt from paying duties (customs tariffs) when they bought railway stock abroad. This was criticised because it could constrain (even impede) the development of the domestic steel industry.

13 See Vidal Olivares and Ortúñez (2002) for further details on the reasons underpinning this increase in ownership.

14 Commercial Code (1829: Art. 36), Regulation (1848, 17-Feb) and Law (1856, 11-Jul).

15 Commercial Code (1829: Art. 36).

16 Regulation (1848, 17-Feb).

17 Regulation (1857, 12-Dic: Art. 1).

18 Royal Order (1864, Sep).

19 Commonly used also are value relevance models. These models study the relation between accounting and prices and returns, with higher correlations being interpreted as indicative of higher quality. In a historical context, their implementation is not always viable, since a strong assumption of these models is that capital markets (and therefore, prices and returns) are efficient and reflect all available information (Holthausen and Watts 2001).

20 NORTE had the following reserves that originated from retained earnings: (1) ‘Reserve for exceptional maintenance and improvement’: this reserve was used to cover the replacement and reinforcement costs of bridges (Annual Report NORTE 1907); (2) ‘Contingency reserves’: these funds were created to ensure the stability of dividends (Annual Report 1910); (3) ‘Funds for the depreciation of material’: these funds were assigned to cover losses when units of material were eliminated from the inventories (Annual Report 1911); and (4) Other Reserves.

21 Building on this sentence, Fidalgo et al. (2015) speculate that it is possible that the ‘Extraordinary expenses and complementary’ account was used to distribute the ‘amortisation’ of these machines to the operating account. These 13 machines were already out of service. Therefore, the substance of the operation was to remove these assets from the accounting, as they were useless, making it dubious evidence of depreciation. In fact, ‘Extraordinary expenses and complementary’ account appears in the income statements of MZA since 1869, when the main lines were built.

22 Considering only years with profits, the median pay-out ratio was 58% from 1920 to 1925, and 97% from 1926 to 1933. This was driven by a dividend which remained fairly stable despite a decreasing trend in earnings. These earnings would have been much lower if NORTE had accounted for depreciation. The average depreciation expense in the 5 years before they stopped the depreciation charge (from 1912 to 1916) was 8,422,435 ptas. Such a charge would have lowered reported earnings by a median of 47.77% from 1920 to 1934.

23 This explains why the persistence model can only be run for 156 observations. The model requires data for t − 1 is available. Data to provide descriptive evidence are available from 1856 for MZA and from 1860 for NORTE, but we have no data for 1855 for MZA and for 1859 and 1867 for NORTE.

24 Spanish railways reported earnings calculated in a uniform manner over time (Santos-Cabalgante et al. 2014).

25 Following the Monetary Reform of 1882, the equivalence was fixed at 1 peseta equals 4 Reales de Vellón. Both firms transition to using PTA in 1882, but this means that MZA uses Rs.Vn. until 1882 (included), whilst NORTE reports in PTA already in 1882.

26 When including the controls for the different phases of the historical context, we can retain all four identification variables in our main analyses (i.e. Start-up, Expansion, Consolidation and Institutionalisation) as well as the intercept as the last of our phases goes from 1919 to 1935, and we have data up to 1939.

27 Using a large sample for the period 1984 to 2004, Dichev and Tang (2009) run earnings persistence models by quintiles of earnings volatility. Their evidence suggests overall high persistence, with persistence coefficients that are on average between 0.51 (for the quintile with the highest volatility) and 0.93 (for the quintile with the lowest volatility). The Adj. R² in their models ranges from 0.296 (highest volatility quintile) to 0.704 (lowest volatility quintile). Although their data cannot be directly compared to ours, for obvious reasons, it is useful in providing a framework to understand the expected values of persistence coefficients and explanatory power of our models.

28 Prior research notes that the crisis of 1883 affected the territories where NORTE and MZA operated differently and that this was reflected in the figures of the two companies. Also, the average prices charged by the two companies before 1898 were different (see Gómez-Mendoza and San Román 2005, p. 513). To control for these potential sources of differences, we conducted a robustness analysis, where we introduce in the model a dummy variable to control for price differences (PRICE_DIF takes the value of 1 before 1898 and 0 otherwise), as well as to control for the crisis of 1883 (CRISIS_1883 takes the value of 1 in 1883 and 0 otherwise). The results, not reported, indicate that our main findings are retained, as our main inferences remain the same when we repeat the analyses of Tables 4 and 5 by company.

29 The report of the first auditory of railway companies criticised NORTE, because it did not record the advance payments as operating expenses (Villacorta Hernández and Müller 2014, p. 158), however, such recording would have lowered earnings, providing much more support for the need for public funds. The fact that NORTE opted not to record these expenses supports the view that the preference was not to minimise earnings.

Additional information

Funding

We acknowledge financial assistance from the Spanish Ministry of Science and Innovation (Ministerio de Ciencia e Innovación) (ECO2013-48328, ECO2016-77579). CAM (H2015/HUM-3353), the Catedra UAM-Auditores Madrid, Spanish Railway Foundation, and 6th ed. AECA Research Grants (2015–2016).