DISCOUNTS remain the most puzzling aspect of investment trusts so far

as most new investors are concerned. Put quite simply, the discount is

the gap between what trusts sell for and what the real worth of their

underlying portfolio actually is.

A trust with 1,000,000 shares might be worth #1m. Each share would

have assets of #1. If it sold for less, it would be said to sell at a

discount; if it sold for more it would be selling at a premium.

Discounts, or premiums, arise because there is no ''right'' price for

any company's shares. Investment trusts are first and foremost

companies. Their business consists in buying, holding, and selling

shares in other companies so as to benefit their own shareholders in two

ways; first, by generating a flow of income from dividends from the

shares they hold, and secondly from enhanced capital values enjoyed by

these shares.

That flow of income is passed on to the trust's own shareholders after

paying the interest on any borrowings the trust has incurred to gear up

the possible capital growth.

How people will value any trusts' shares depends on an assessment of

these factors. How good has a management been at making the assets grow,

and using borrowing prudently? How good at increasing dividends, in

spite of paying the interest on borrowings?

These, and similar, assessments have always affected the rating of

investment trust shares but one vital factor reduced the impact: there

were far more investment trust shares available than there were

shareholders willing to take them.

The discount mirrored that fact. The law of supply and demand

dominated the valuation of investment trust shares as it does the shares

of any company. Professional investors traded the shares within limits

laid down by size of the discount which widened and narrowed

accordingly.

Investors were always unhappy about discounts because although they

delivered compensations (a higher yield than would otherwise have been

provided, for example) they added to trusts' price volatility. Managers

were affected by discounts as well. They were not good for morale, and

they greatly reduced the ability to raise new capital or start new

investment trusts.

Two forces have moved to eliminate the discount. The first is the

changed climate for private investors, the second the ingenuity of

increasingly professional and sophisticated trust managers.

Although private investors have continued to be net sellers of

equities during the past decade, they have become big buyers of

investment trust shares. Helped by fiscal changes, British companies

have been paying higher dividends since 1979 and trusts have started to

support healthy and growing yields which are worth much more to

recipients.

Trusts also partly ran down their gearing, which has left more of

their income for dividend payments.

Savings schemes have started to do the trick of returning trust shares

to those for whom they were originally intended -- individual, and not

institutional, holders.

Apart from the effect of new and sustainable demand from private

investors, the main force tending to reduce discounts has been

structural. Clever managers noticed that, if they packaged their trusts

in novel ways, the valuations placed on the various parts of the package

would collectively eliminate the discount.

That is to say, if you launched a trust with more than one category of

share the valuation placed on each category, when added up, would come

to near, and possibly even more than, the total assets the trust would

enjoy.

The two essential features in this process were options and split

levels. Options, known as warrants, permit the future purchase of a new

share in the trust at the issue price. The normal structure has one

warrant for ever five shares initially issued.

The future usually runs for up to 10 years, with the warrant owner

being able to exercise his option annually. That right to buy is always

worth something; and that something was often enough to allow trusts to

be launched without going to a discount of much significance.

Since there is nothing for nothing, the sting in the tail of every

warrant is its dilution of the existing capital when exercised. But the

exercise date is far away when a trust, or a warrant, is launched. In

the shorter term the existence of the warrant prevents a big discount

from opening up.

The second device was to split the trust's capital in such a way that

some shares got all the income and some all the capital growth over a

specified lifetime, seldom more than 20 years and sometimes less. There

could be variations on this theme whereby some capital growth would be

available to the income shares (an example is River and Mercantile

Geared).

Getting all the income from both sorts of capital would mean that

income shares would offer such a large, and growing, dividend return

that they would retain their value (or go to a modest premium in the

package).

The capital shares would normally trade at a discount, but that was

eliminated by an immediate transfer of assets from the income stocks

which would remain floating high because of the high dividends promised.

In effect income shares boost the capital shares from the beginning of a

trust's life.

Such structures have become ever more complex with zero pref-erence

capital shares being added. These enjoy no income but a guaranteed

annual capital increment payable when the trust is wound up. They help

therefore to gear up the ordinary capital shares and, eventually, boost

the income of the income shares.

Where warrants are added, allowing future purchase of geared capital

shares, the result is a very highly geared instrument indeed (viz

Scottish National Trust warrants.)

When all instruments are lumped together at their market valuations,

hey presto, the discounts have either disappeared or become negligible;

and a important psychological barrier to buying investment trusts has

been dismantled.