RDP 2016-03: Why Do Companies Hold Cash? 4. Data
May 2016
- Download the Paper 1.07MB
To examine corporate cash holding behaviour we utilise two unique sources of Australian company-level panel data that cover both public and private companies and are provided by D&B and Morningstar.
4.1 The Dun and Bradstreet Database
In the D&B data there are around 4,000 private companies and 1,400 unlisted public companies (those with shares that are not traded on a public exchange), and the annual data cover a 10-year sample window from 2005 to 2014.[18]
The D&B sample of companies is based on its primary business as a credit bureau. Companies in the database are likely to be those that apply for credit, and D&B can most easily obtain financial information on relatively large private and unlisted public companies, which are required to file their financial reports with the Australian Securities and Investment Commission (Table 1). As a result, compared to the population of all private and unlisted public companies in Australia, the sample is biased towards larger companies, though it does still have some coverage of very small companies (those with revenue less than $0.5 million).
The slant towards larger companies that want to borrow does not necessarily undermine the motivation for our work. In fact, it is not clear how this will bias our results (see Appendix C). Moreover, the selection criteria and variables of interest included in the database are consistent across public and private companies, which allows us to make direct comparisons.
In the unbalanced D&B panel of private and public companies, about 75 per cent of companies are observed for two years or more; 50 per cent of companies are observed for three years or more; and 25 per cent of companies are observed for five years or more.
Summary statistics for the D&B data are presented in the first two blocks of Table 2. Unlisted public companies are smaller than private companies, on average, though for both groups of companies holdings of total assets are unevenly distributed, with the mean level of assets above the 75th percentile of the distribution.
Table 2 shows that cash ratios are higher for unlisted public companies compared to private companies and that unlisted public companies tend to be older than private companies. Likewise, capital spending as a share of total assets is generally larger for unlisted public companies.
The ratio of cash flows (or earnings) to assets is quite similar across public and private companies.[19] Cash flows are more volatile for private companies as measured by the standard deviation in company-level cash flows (averaged across industries). The ratio of working capital (measured as the stock of inventories and short-term receivables outstanding) to assets is substantially larger for private companies. Finally, private companies are more leveraged as they are more reliant on debt, while public companies are able to tap public equity markets.
| Mean | 25th percentile | Median | 75th percentile | |
|---|---|---|---|---|
| Private D&B companies: 2005 to 2014 | ||||
| Real total assets ($m)(a) | 80.1 | 14.8 | 27.2 | 63.4 |
| Cash ratio (%) | 11.4 | 1.8 | 6.5 | 15.9 |
| Age (years) | 23 | 11 | 19 | 30 |
| Capex ratio (%) | 3.8 | 0.7 | 1.9 | 4.6 |
| Cash flow ratio (%) | 8.9 | 1.1 | 8.0 | 16.3 |
| Risk (ppt) | 18.4 | 9.3 | 13.2 | 18.3 |
| Working capital ratio (%) | 46.2 | 24.6 | 46.9 | 68.2 |
| Leverage ratio (%) | 62.6 | 40.7 | 60.3 | 79.9 |
| Number of observations | 14,646 | |||
| Number of companies | 4,085 | |||
| Unlisted public D&B companies: 2005 to 2014 | ||||
| Real total assets ($m)(a) | 70.3 | 6.3 | 15.8 | 43.1 |
| Cash ratio (%) | 16.3 | 3.3 | 8.6 | 21.7 |
| Age (years) | 35 | 18 | 34 | 43 |
| Capex ratio (%) | 6.6 | 1.8 | 4.4 | 8.8 |
| Cash flow ratio (%) | 7.4 | 2.5 | 6.9 | 12.0 |
| Risk (ppt) | 13.7 | 7.0 | 9.1 | 13.9 |
| Working capital ratio (%) | 12.7 | 1.3 | 3.5 | 14.1 |
| Leverage ratio (%) | 40.7 | 19.7 | 34.6 | 55.2 |
| Number of observations | 5,038 | |||
| Number of companies | 1,402 | |||
| Listed public Morningstar companies: 1990 to 2014 | ||||
| Real total assets ($m)(a) | 623.2 | 8.0 | 23.5 | 104.6 |
| Cash ratio (%) | 26.2 | 4.4 | 14.4 | 40.7 |
| Age (years) | 17 | 6 | 12 | 23 |
| Capex ratio (%) | 5.3 | 0.3 | 2.0 | 6.6 |
| Cash flow ratio (%) | −7.1 | −10.3 | −1.5 | 5.7 |
| Risk (ppt) | 12.8 | 7.0 | 12.5 | 15.2 |
| TQ (ratio)(b) | 2.1 | 0.8 | 1.2 | 2.1 |
| Research and development ratio (%) | 8.1 | 0.1 | 0.9 | 6.5 |
| Working capital ratio (%) | −3.6 | −6.9 | −1.1 | 5.9 |
| Leverage ratio (%) | 31.1 | 5.0 | 20.1 | 45.1 |
| Number of observations | 23,562 | |||
| Number of companies | 1,954 | |||
|
Notes: (a) June 2013 dollars Sources: Authors' calculations; D&B; Morningstar |
||||
4.2 The Morningstar Database
For the longer-run analysis, we use detailed financial statement information for Australian publicly listed companies provided by Morningstar. This database covers a narrower sample of companies (at around 2,000 companies) than the D&B database, but a much longer sample window of 1990–2014. This allows us to better gauge the determinants of the long-run trend increase in corporate cash holdings.
About 75 per cent of companies are observed for 7 years or more; 50 per cent of companies are observed for 11 years or more; and 25 per cent of companies are observed for 17 years or more.
While it is difficult to make direct comparisons between the listed public companies and unlisted public companies as some of the data definitions vary across databases (see Appendix C), generally speaking listed public companies are much larger than their unlisted counterparts (Table 2, third block). The cash ratios of listed public companies are also larger than unlisted public companies.
Footnotes
In this paper, financial companies are excluded because they may carry cash to meet capital requirements that are unrelated to the reasons investigated in this paper. Likewise, utilities companies are dropped because their cash holdings – in part – may be determined by regulations. In most cases the data used are financial year-ended; however, in some instances companies report calendar year-ended figures, or annual figures ended in other months. [18]
It may seem surprising that the cash flow-to-assets ratio for listed public companies is negative, on average. However, this appears to reflect a relatively large share of Australian listed companies that are small mining exploration companies that typically report losses. The high share of companies reporting losses can also be observed in Australian Taxation Office data on company profitability. [19]