Impaired Loss Accounting Does Not Affect Firms’ Investment

Impaired loss accounting has become a familiar term due to media reports about Toshiba Corporation's business crisis. Toshiba incurred huge losses as it reported impaired losses on goodwill that arose after the acquisition of Westinghouse Electric Company of the United States and is being forced to sell its semiconductor division (See Box: Impaired Loss Accounting and Goodwill).

The fall from grace of this major Japanese electronics company may have deepened the impression that impaired loss accounting is the culprit of the Toshiba crisis. However, in fact, since before the introduction of the mandatory application of the impaired loss accounting standard in Japan in FY2005, it has been pointed out that reporting of impaired losses on fixed assets may have some negative impact on companies' performance or activities.

Around the time of the introduction, the focus of concern was impaired losses on tangible fixed assets, such as land and buildings, rather than on goodwill. The possibility was pointed out that vast amounts of unrealized losses on real estate properties that existed since the collapse of the asset price bubble would be revealed through impaired loss accounting, and, as a result, it could become difficult for companies to raise funds due to a decrease in the value of tangible fixed assets that can be used as collateral or a decline in revenue.

It is important to find out whether that risk has been realized and whether impaired loss accounting has affected companies' activities in other ways if we are to consider the role of impaired loss accounting now that Toshiba is teetering on the brink. This article will report on what changes impaired loss accounting has brought hitherto to Japanese listed companies' capital investment and land transactions. (For details, see Uesugi, Iichiro, Kentaro Nakajima, Kaoru Hosono, "Does Impaired Loss Accounting Affect Firms' Investment? Evidence from Japan" RIETI Discussion Paper Series 17-J-033).

After the collapse of the economic bubble, it was pointed out that the significant decline in the value of assets held by Japanese companies was not disclosed in a timely manner in their financial statements. In light of this, in August 2002, the Business Accounting Council announced the "Accounting Standard concerning Fixed Asset Impaired Losses" in line with the international trend of the development of accounting standards intended to reflect fixed asset impaired losses, thereby paving the way for the introduction of impaired loss accounting in Japan. Some politicians and companies were concerned that the application of the impaired loss accounting standard would increase deflationary pressure by revealing unrealized losses on corporate assets. However, the accounting standard was introduced in FY2003 in early application cases as initially planned.

Following the introduction of the impaired loss accounting standard, a fairly large number of companies reported impaired losses on tangible fixed assets. Table 1 shows the proportion of companies which reported impaired losses among the companies covered by our analysis.

In FY2003 and FY2004, immediately after the start of the application of impaired loss accounting in early application cases, the proportion of companies which reported impaired losses was small but increased significantly in FY2005, when the mandatory application started. According to a survey conducted by the Policy Research Institute of the Ministry of Finance, which conducted a population estimate concerning Japanese companies as a whole, impaired losses reported over the first three years after the introduction totaled approximately 6.9 trillion yen. The amount of reported impaired losses has stayed at a high level since FY2006. This suggests that in addition to losses caused by falls in real estate asset prices after the collapse of the economic bubble, a large amount of losses arose after the application of the impaired loss accounting standard.

In addition to the proportion of companies which reported impaired losses on tangible fixed assets (center column), Table 1 shows the proportion of companies which reported impaired losses on overall fixed assets (left column), including intangible assets, and the proportion of companies which reported impaired losses on intangible fixed assets (right column). In most cases, impaired losses were reported on tangible fixed assets, such as land and buildings, while there were relatively few cases of impaired losses reported on intangible fixed assets such as goodwill as in the case of Toshiba.

How could companies' activities be affected by reporting of impaired losses, mainly on tangible fixed assets, by a fairly large number of listed companies? There are three hypotheses concerning possible impacts in this regard.

The first hypothesis is that there will be a negative impact, as mentioned earlier. When companies raise funds from financial institutions with tangible fixed assets as collateral, if impaired losses are reported based on a market price of the fixed assets lower than the book value, their borrowings are constrained by a decline in the value of the collateral assets. As a result, investment will decrease. The impact appears not only in the form of a decline in the value of collateral assets. Reporting of impaired losses also lowers companies' profits, thereby strengthening constraints on their borrowings and reducing their investments.

The second hypothesis is that there will be a positive impact. If companies report impaired losses at an early stage, that will have the effect of increasing the credibility and transparency of financial information disclosed by them. If this effect is strong, it will become easier for companies struggling with the problem of underinvestment to borrow funds. Indeed, according to a study conducted by Garcia Lara et al. (2016), which used U.S. data, there is a tendency that companies reflecting losses in their accounting at an early stage can avoid the problem of underinvestment despite facing financial constraints.

Although these two hypotheses predict opposite impacts, both are based on the assumption that companies' reporting of impaired losses provides new information for financial institutions and capital markets that supply funds.

On the other hand, the third hypothesis assumes that as financial institutions and capital markets are already aware of companies' unrealized profits and losses, reporting of impaired losses does not provide any new information. According to this assumption, the fund supply side constantly keeps track of the value of fixed assets held by companies based on disclosed information, so losses foreseeable from the information have already been taken into consideration. Therefore, if companies report impaired losses, that does not have a significant impact on companies' fundraising or investment.

In order to judge which of the three hypotheses is correct, it is necessary to conduct a quantitative analysis using actual data. For this purpose, we employed firm-level data of about 850 companies which have continuously responded to the Ministry of Finance's Surveys for the Financial Statements Statistics of Corporations by Industry since FY1980.

We examined how companies' activities, such as capital investment and land transactions, are affected by the presence or absence of not only reporting fixed asset impaired losses but also unrealized profits on land holdings (= market value of land held−book value) after the reporting of impaired losses. Unrealized profits on land holdings can be calculated based on the history of land purchases by individual companies as recorded in statistical data.

If the first hypothesis is correct, capital investment and land purchases by companies which reported fixed asset impaired losses are expected to decrease, while if the second hypothesis is correct, such expenditures are expected to increase. If the third hypothesis is correct, capital investment and land purchases by companies with a large amount of unrealized profits on land holdings are expected to increase, whereas the presence or absence of reporting of impaired losses is not expected to have a significant impact on those variables.

The result suggests that reporting of impairment losses does not have a significant impact on capital investment or land purchases. Neither the capital investment ratio nor the land purchase ratio changed significantly (Table 2).

Separately from the above analysis, we also conducted estimation concerning impaired losses reported on intangible assets such as goodwill as in the case of Toshiba but did not find a significant impact on investment by companies which reported such losses.

Another notable result was that an increase in unrealized profits on land holdings that arose after reporting of impaired losses had a statistically positive correlation with the capital investment ratio. According to column (i) of Table 2, the average company makes a capital investment equivalent to 7.7% of the capital stock balance per year, while the capital investment ratio of companies with a relatively large amount of unrealized profits on land holdings is 0.7 percentage points higher. In other words, the presence of a relatively large amount of unrealized profits on land holdings has the effect of increasing capital investment by nearly 10% compared with the average.

These results do not support either the hypothesis that reporting of impaired losses damages the investment environment by reducing the value of collateral assets or the hypothesis that it makes fundraising easier by increasing the transparency and credibility of financial information. However, it should be noted that the analysis does not cover companies listed on emerging market exchanges and small and medium-size enterprises, which tend to face financial constraints. Therefore, a negative impact from reporting of impaired losses may be observed if we include these firms in the analysis.

On the other hand, the results suggest the possible viability of the hypothesis that reporting of impaired losses does not provide the kind of information that would cause changes in corporate investment by changing fund suppliers' activities. It is possible that this hypothesis is also viable with respect to reporting of impaired losses on intangible assets although it may appear so because of the small sample size.

Although the impact of reporting of impaired losses on capital investment and land purchases is not clearly visible, companies which reported impaired losses recorded a relatively high land sales ratio in the same year, according to column (iii) of Table 2. The average company sold land equivalent in value to 1.8% of the land stock balance per year, whereas the land sales ratio of companies which reported impaired losses was 0.6 percentage points higher. In other words, the presence of impaired losses has the effect of increasing land sales by just over 30% compared with the average.

According to column (iii), among companies which reported impaired losses, there was a tendency that strongly performing ones with rising profitability conducted a relatively large amount of land sales. This indicates that companies which reported impaired losses sold land due to motives other than obtaining liquidity in the face of financial constraints.

One presumed factor that encourages companies which reported impaired losses to sell land is a desire to eliminate the difference between their profits as defined for the accounting purpose and their income as defined for the tax payment purpose. Even if companies write down the value of fixed assets or decrease profits calculated for the accounting purpose based on the impaired loss accounting standard, the impaired losses are not included in losses as calculated for the tax payment purpose under the current Japanese tax codes. This means that there will be no change in the amount of corporate tax. Therefore, companies which earn large income and thus likely to face a high corporate tax bill may sell fixed assets at the same time as writing down the value of the assets in order to include the fixed asset impaired losses calculated for the accounting purpose in the losses calculated for the tax payment purpose. The results shown in column (iii) indicate that companies with rising profitability tend to sell land at the same time as reporting impaired losses because they are strongly motivated to increase the amount of losses calculated for tax payment purposes.

We examined the impact of reporting of fixed asset impaired losses by companies on their investment. At the time of the introduction of the impaired loss accounting standard, there were some concerns that reporting of impaired losses on fixed assets might lead to a decrease in investment by impeding companies' fundraising. But such a negative impact was not observed, at least among the listed companies that were covered by this analysis. Although there are multiple reasons for this result, it is possible that ordinary reporting of impaired losses did not necessarily provide new information for the capital supply side such as investors and financial institutions.

What has become statistically evident is the tendency of companies which reported impaired losses to sell land holdings. The companies sold land holdings, rather than maintaining them, presumably in order to reflect a decline in their profits for the accounting purpose caused by impaired losses in the income for tax payment purposes. The Japanese real estate market and the resource allocation between companies may have been significantly impacted over the so-called lost two decades by the fact that companies sold land holdings after reporting impaired losses because of the difference between the profits defined for accounting and tax administration purposes. Whether the impact was positive or negative is a matter for future examination.

>> Original text in Japanese

* Translated by RIETI.