In 1999 Kotsovolos, the leading Electronics Supplier in Greece, reported in its initial public offering prospectus an earnings forecast that missed its actual earnings, as announced by its first annual report, by 234%. This inaccuracy is attributed to the mandatory disclosure requirement imposed by the Hellenic Capital Market Commission, which obligated every firm going through an IPO to predict its next year’s earnings regardless of its ability to do so. Ultimately, repeated failures to achieve accurate earnings forecasts led to a lifting of the obligation to forecast earnings. Our new paper, Voluntary vs Mandatory Management Earnings Forecasts in IPOs, explores this regulatory change and provides empirical evidence of the accuracy of earnings forecasts in both mandatory and voluntary disclosure environments.

The change in regulation took place back in 2000 when the authorities realized the extent of the failure of firms to provide accurate earnings forecast in their IPO prospectuses. The problem of inaccurate earnings forecast was exacerbated during the late 1990’s when Greece had a large influx of newly listed companies on the Athens Stock Exchange. This influx was largely driven by the global technology sector boom, which resulted in the listing of companies with minimum prior experience but great expectations regarding future corporate earnings and growth prospects. Our findings indicate that those companies not only failed to provide an accurate earnings forecasts but contributed to a deteriorating level of trust between management and investors.

The danger of market scorn due to faulty management forecasts as well as the complaints of newly listed firms concerning compliance costs and their inability to provide accurate management earnings forecasts prompted the Capital Market Commission to remove the mandatory earnings forecast requirement. The regulatory switch from mandatory to voluntary management earnings forecasts was further motivated by its supposed contribution to the efficient and cost-effective functioning of the capital market. The protection of investors and the maintenance of confidence in the Greek financial market were related important objectives. This regulatory change was also intended to reinforce the freedom of movement of capital in the internal market and to incentivize small family companies to go public.

Although managers do not entirely control the news that their forecast convey under the voluntary disclosure environment, managers can effectively create such control within the current period via their decisions about whether to issue an earnings forecast and the forecast earnings number itself. The news conveyed by a forecast falls into one of three categories: good-news forecasts exceed market expectations; bad-news forecasts fall short of these expectations; and confirming forecasts corroborate those expectations. In many cases, researchers use analysts’ consensus earnings estimates as a proxy for market earnings expectations. Our research shows a changing trend over time in terms of management earnings forecast news, (while international evidence indicates that good-news forecasts and bad-news forecasts are equally likely in the studied time period).

While 100% of Greek IPOs issue forecasts during the mandatory forecast regime, only 70.9% of firms issued forecasts during the voluntary regime. This is in contrast to U.S. firms who are less likely to provide earnings forecast in the corresponding U.S. voluntary disclosure environment. We find that following the introduction of the voluntary reporting regime, the average age of Greek firms going public and choosing not to provide earnings forecasts is 5.92 years. In contrast, the average age of companies that voluntarily provide a forecast is 17.29 years, thereby providing an initial signal that young, immature companies choose to omit this information.

Breaking down the forecast error by mandatory and voluntary environments, the results reveal a positive mean of 8.65% for IPOs providing an earnings forecast during the mandatory period and a negative mean of -9.58% for IPOs providing a forecast of earnings during the voluntary period. This indicates that firms are very conservative when obliged to provide a forecast and therefore forecast profit is lower than actual. Once the disclosure environment turns to voluntary, Greek IPOs turn more liberal and the forecast error sign typically changes indicating more optimistic forecasts. This supports to the self-selection theory, which in part states that earnings forecasts during voluntary periods are generally overoptimistic

To capture the effect of the regulatory change, we introduce a ‘mandatory’ variable, which separates those IPOs that were forced to provide earnings forecast during the listing period from those that voluntarily decided to announce their expected earnings. Results show that the mandatory coefficient is positive and statistically significant. Further, we follow the same analysis for forecast error and the coefficient on the ‘mandatory’ variable is again positive and significant. Thus, our results imply that the disclosure of a forecast of earnings in a mandatory regime provides more – but not necessarily more useful – information to investors as the accuracy of the information is suspect.

In order to shed more light on the relationship between the regulatory change and earnings forecast accuracy, we investigate the impact of other regulatory changes that took place during the studied period. Globally, the Athens Stock Exchange has been one of the few stock exchanges that introduced a price cap during the first day of trading after listing. Specifically, there has been a price cap of ± 8% during 1993-1996, changing to ±99% during 1996-1999 and no price cap thereafter. We start by testing the price cap period to explain the level of forecast error FE and we test for the consecutive limit number of up ticks in the stock price CLU. The results indicate that the coefficients for the price cap and consecutive limit ups are positive and significant at the 10% level. In economic terms, going public with a restriction on daily variation increases the level of error of the earnings forecast during the mandatory disclosure period. Price caps appear to make estimates more conservative as firms going public feel uncomfortable about providing an optimistic earnings forecast. We also examine the change of pricing mechanism from fixed offer price to bookbuilding and the relation with the management earnings forecast. Findings show that the bookbuilding mechanism (i.e. underwriters solicit investor information as part of pricing and, to create incentives for investors to be truthful, they deliberately underprice offerings and give allocation priority to investors who reveal strong interest) is positive and significant at the 1% level for absolute forecast error. That is, companies which select the book building mechanism (i.e process of generating, capturing, and recording investor demand for shares during an initial public offering in order to support efficient price discovery) when going public experience larger forecast error.

A natural question that arises from our findings is why there is a change in the bias of earnings forecasts following the reform from mandatory to voluntary disclosure of an earnings forecast in the prospectus. Our view of the management behavioural change is that, initially, it is aftermarket uncertainty surrounding the company’s future operations that leads many issuers to understate their earnings forecasts during the mandatory regime and to provide conservative (pessimistic) earnings forecasts Then, following the regulatory change, those managers facing uncertain future operations choose not to disclose earnings forecast during the voluntary period thereby allowing more optimistic views to dominate.

These findings led us to offer the following suggestions that would make management earnings forecast more efficient and bring their disclosure into closer conformity with investors’ interest.

1. The Capital Market Commission should introduce specific standards governing whether a company may voluntarily provide earnings forecasts in its IPO prospectus. This will automatically rule out all those companies which are unable to provide such a forecast as there will be certain restrictions inspired by historical accounts, leverage, amount of gross proceeds and structure of use of capital raised. Those limitations will contribute on eliminating specific firms by providing earnings forecast or by being overoptimistic.
2. Prepare an error metric of previous management earnings forecasts that provides information to investors to mitigate the gradual spillover of the adverse effects to the real economy, and the perpetuation of a negative investment climate.
3. Inform firms interested in a public offering about previous IPO experience to protect them from mistakes that have taken place in the past. Towards this the IPOs and their management that achieve high level of accuracy should be publicly noted and rewarded for the quality they deliver to the market.

We believe that Exchange Commissions can implement these reforms, as they can help achieve better operations of the capital markets which are of great interest to the investors.

Dimitrios Gounopoulos is a Reader in Accounting and Finance at the School of Business, Management and Economics at the University of Sussex at Brighton, UK. Arthur Kraft is a Senior Lecturer in Accounting at City University, London. Frank Skinner is a Professor in Finance at the School of Finance and Economics at Brunel University in London. This post is based on their recent article, which is entitled “Voluntary vs. Mandatory Management Earnings Forecasts in IPOs” and available here.