Section 42 of the Income Tax Act 58 of 1962 (Act) deals with asset-for-share transactions. It provides roll-over relief for persons wishing to transfer assets to a company in exchange for an issue of shares. But how many shares must the transferee company issue in exchange for an asset or assets?
Before attempting to answer this question, it might be helpful to explain the basics of s 42 with reference to a simple example.
Example 1 - Basic asset-for-share transaction
Facts:
John owns a piece of vacant land which he acquired on 1 March 2008 at a cost of ZAR 20. The current market value of the land is ZAR 100. He wishes to transfer the land to Newco (Pty) Ltd (Newco), a resident, in return for shares in Newco. John held the land as a capital asset and it is to be acquired by Newco as a capital asset. Newco will issue 100 equity shares with a value of ZAR 100 in exchange for the land. John will be the sole shareholder of Newco after the transaction.
Result:
John is deemed to dispose of the land for proceeds equal to its base cost (s 42(2)(a)(i)(aa) read with para (a) of the definition of ‘asset-for-share transaction’ in s 42(1)) and so will realise neither a capital gain nor a capital loss (ZAR 20 – ZAR 20). The market value of the land exceeds its base cost (ZAR 100 > ZAR 20) on the date of disposal and John holds a qualifying interest in Newco (at least 10% of the equity shares and voting rights) (para (a)(i) of the definition of ‘asset-for-share transaction read with the definition of ‘qualifying interest’). Under s 42(2)(a)(ii) John is treated as having acquired the 100 equity shares on 1 March 2008 at a cost of ZAR 20 incurred on 1 March 2008 (this aspect is the focus of this article). Under s 42(2)(b) Newco is treated as having acquired the land on 1 March 2008 at a cost of ZAR 20 incurred on 1 March 2008. |
So far so good. John sold an asset to Newco at no gain/no loss and acquired 100 equity shares which have taken over the characteristics of the land in relation to cost, date of acquisition and date of incurral of expenditure. If he sold the shares the next day for ZAR 100, he would realise a capital gain of ZAR 80, which is the same result that would have ensued had he sold the land for ZAR 100.
But what happens when there are multiple assets disposed of in exchange for shares?
Example 2 - Multiple assets disposed of under an asset-for-share transaction
Facts:
Jill owns two post-valuation date capital assets which she wants to dispose of to Newco (Pty) Ltd (Newco) under an asset-for -share transaction in exchange for two shares. The base cost of each asset is equal to its market value. Asset A’s base cost is ZAR 100 and Asset B’s base cost is ZAR 900.
Result:
The problem that arises here is that if share 1 is allocated to Asset X and share 2 to Asset Y, there will be a distortion between the base cost of each share and its market value. Assuming this is the only transaction, each share should be worth ZAR 500 (ZAR 100 + ZAR 900 = ZAR 1 000/2 = ZAR 500). If Jill were to dispose of each share the next day, there would be a capital gain of R400 on share 1 (ZAR 500 – ZAR 100) and a capital loss of ZAR 400 on share 2 (ZAR 500 – ZAR 900). If both shares are disposed of simultaneously, it would not be an issue because the aggregate base cost of both shares would equal their market value and the gain on Asset X could be set off against the loss on Asset Y, assuming the loss is not clogged under para 39. |
So, what does s 42 actually require? The definition of 'asset-for-share transaction'
‘means any transaction … in terms of which a person disposes of an asset … in exchange for the issue of an equity share …’.
The definition, if taken literally, requires a single share to be issued for each asset.
Section 42(2)(a)(ii) on the other hand treats the transferor to have
Thus, in this instance the provision contemplates multiple shares for a single asset.
Section 6(b) of the Interpretation Act 33 of 1957 states the following:
'In every law, unless the contrary intention appears … words in the singular number include the plural, and words in the plural number include the singular.'
Given the conflicting use of the singular in the definition of ‘asset-for-share transaction’ and the plural in s 42(2)(a)(ii), it would be reasonable to infer that one should not attach too much importance to the use of the plural and singular in s 42 when it comes to how many shares are required to be issued. As long as at least one share is issued for an asset or assets, the requirements of s 42 in relation to the issue of a share or shares should be met.
But this does not solve the distortion problem that arises when there are insufficient shares to match the relative market values of the assets forming part of the subject of an ‘asset-for-share transaction’.
One way in which to solve the problem would be to issue a sufficient number of shares so that the base cost of an asset can be matched with the required number of shares. Thus, in Example 2, Newco could have issued 1 000 shares for R1 000, and allocated 100 shares to Asset X and 900 shares to Asset Y.
However, sometimes a company may not wish to issue so many shares, particularly when there are minority shareholders, as this may upset the balance of control over the company.
Another approach to this problem would be to simply allocate the aggregate base cost of all the assets to the shares issued, so that in Example 2 each share would have a base cost of ZAR 500.
There is in my view, much to commend this approach as it avoids the problem of maintaining a share register in which different base costs are allocated to certain batches of shares with distinctive certificate numbers. One can only imagine the nightmare involved in tracking shares linked to thousands of assets.
One cannot help being reminded of the oft-cited passage from
Natal Joint Municipal Pension Fund v Endumeni Municipality in which the court stated the following:1
'An interpretation will not be given that leads to impractical, unbusinesslike or oppressive consequences or that will stultify the broader operation of the legislation or contract under consideration.'
Pre-valuation date assets
While the aggregate base cost allocation method works well for post-valuation date assets, it does not work for pre-valuation date assets.
The problem with pre-valuation date assets is that the shares to which they are linked will also become pre-valuation date assets. There are three methods for determining the valuation date value of a pre-valuation date asset, namely, market value on 1 October 2001, the time-apportionment base cost method and the ‘20% of proceeds’ method. If there is a record of pre-valuation date expenditure, the time-apportionment base cost method can be used.2 This method requires the taxpayer to know the date of acquisition, the para 20 expenditure and the date of incurral of the expenditure. It is highly unlikely that all the assets that are the subject of a s 42 transaction would have the same dates of acquisition and incurral and cost. In addition, the problem with time-apportionment is that it represents a constantly moving target when it comes to the determination of the valuation date value of an asset at any particular point in time. ‘T’ in the time-apportionment formula3 (the number of years or part thereof on or after valuation date) keeps changing the longer the asset is held. Generally, the longer the asset is held after valuation date, the greater the proportion of the capital gain or loss that will have to be brought to account. What also complicates matters is that some assets involved in an asset-for-share transaction may have been improved after the valuation date, which triggers the proceeds formula4 or the depreciable assets formula.5 And when an asset has been improved in more than one year before the valuation date, ‘N’ (the number of years or part thereof before 1 October 2001), is limited to 20.6
Aggregating costs in these circumstances is simply not an option.
The solution to this conundrum is to allocate shares with distinctive certificate numbers to particular pre-valuation date assets based on their relative market values.
Another solution to this problem would require legislative intervention. A rule similar to that found in para 76B(1) of the Eighth Schedule to the Act could be introduced, which would apply immediately before any asset-for-share transaction is entered into. Paragraph 76B(1) applies when a return of capital is received or accrued on a pre-valuation date share for the first time on or after 1 April 2012. Under this provision, the share must be converted from a pre-valuation date share having a valuation date value to a post-valuation date share having para 20 expenditure. This task is achieved by treating the share as having been disposed of and reacquired at market value. Any capital loss resulting from the deemed disposal is added to the reacquisition cost, while any capital gain reduces it. The capital gain or loss arising under para 76B(1) is purely for the purposes of re-establishing the base cost and is not actually brought to account as a capital gain or loss in determining the shareholder’s aggregate capital gain or loss for the year of assessment.
Of course, a company could solve the problem itself by simply distributing a small amount of contributed tax capital to its shareholder, as this would trigger para 76B(1) and thus convert all the pre-valuation date shares to post-valuation date shares.
Introducing a similar rule under s 42 would assist in simplifying the roll-over of the base cost of the asset to the shares issued by the transferee company. While it may be detrimental to the fiscus to freeze the base cost of a pre-valuation date share whose price is expected to rise, it would prejudice the taxpayer if the company were expected to make losses. However, the benefits of simplification are likely to outweigh any prejudice to the
fiscus or the taxpayer.
Value for value
While ss 41 to 47 generally override the rest of the Act, s 41(2) specifically provides that this does not apply to s 24BA and s 40CA(b). Under s 24BA the value of the asset transferred must equal the value of the shares received in exchange immediately after their issue, otherwise this can result in a capital gain or dividends tax for the transferee company. Nevertheless, s 24BA(3) provides that s 24BA will not apply if
- The transferor and transferee company are part of the same group of companies immediately after the company acquires the asset;
- The transferor holds all the shares in the company immediately after the company acquires the asset; or
- Para 38 of the Eighth Schedule applies.
The sale agreement should specify the consideration for the sale as being equal to the market value of the shares to be received and not the base cost of the assets being transferred.
Conclusion
The allocation of the base cost of post-valuation date assets to shares issued under s 42 on an aggregate basis offers a simple solution to the question of how many shares need to be issued. The decision can be left to practical commercial considerations in the knowledge that the aggregate basis removes any distortion of the base cost of the shares. It would give taxpayers some comfort if SARS were to issue an interpretation note on this subject.
However, when pre-valuation date assets are involved, some burdensome record-keeping may be required. It may be time for some legislative intervention to simplify matters.
The value-for-value rule in s 24BA must always be borne in mind when performing an asset-for-share transaction under s 42.
This article was first published in
ASA February 2024
