Dreistufige Pyramide der Fair-Value-Bewertung: unten Level 1 mit direkten Börsenkursen, in der Mitte Level 2 mit abgeleiteten Werten aus vergleichbaren Marktdaten, oben Level 3 mit eigenen Schätzungen ohne Marktdaten; daneben Pfeile, die zunehmende Unsicherheit nach oben anzeigen.

Fair Value

The fair value is the price an asset would fetch today in a sale between two independent, well-informed parties. It is the central valuation metric in modern balance sheets and there often replaces the historical purchase price.

Fair value is an answer to a simple question: What is a thing worth today? What is meant is the amount one would receive upon selling it on the market. Both sides are supposed to act voluntarily, be independent of one another, and know the item well. No one may be under time pressure or forced into a distress sale. The German technical term for this is “beizulegender Zeitwert.” It stands in contrast to the acquisition cost, that is, the price one paid oneself at some point in the past.

Why companies don’t stick with the purchase price

A balance sheet is a list of what a company owns and owes. For a long time, it mainly listed acquisition costs. A block of shares bought in 2015 would then remain on the books at the 2015 price, regardless of what had happened since. That is very reliable, but it can diverge massively from reality.

With fair value, revaluation happens regularly instead. If the price rises, the balance sheet value rises too. If it falls, the value falls as well. Investors thereby see a current picture instead of a historical one. This is precisely why international accounting standards such as IFRS mandate this kind of valuation for many financial assets.

The price for this is volatility. Balance sheets become more turbulent because market movements feed through directly. During the 2008 financial crisis, this was a hotly debated issue: as prices collapsed, banks had to write down their holdings, which triggered further sales and pushed prices down even more. Critics spoke of a downward spiral.

The three levels of valuation

The simplest case is when there is an open market. A share of SAP has a price every trading day. You simply take that price and you’re done. Experts call this level one, or Level 1.

Often, though, there is no direct price. In that case, one looks for proxy measures: prices of similar products, interest rates, exchange rates. The value is calculated from this observable data. This is level two. An example would be a bond that is rarely traded but whose twin is traded regularly.

Level three is the most difficult. Here, no usable market data exists anymore, such as with a stake in a young start-up. The company then has to estimate the value itself, usually via a forecast of future profits. Such estimates depend heavily on assumptions. If you slightly change the expected growth rate, the result often changes dramatically. That is why auditors and investors scrutinize Level 3 valuations especially closely.

Fair value in tech news and portfolios

In the tech sector, the term comes up when corporations hold stakes in other companies. If a corporation invests in an AI start-up and that start-up is later valued more highly, this can appear as a profit in the quarterly report even though not a single share was sold. Such paper gains are genuinely booked but do not exist as money in the bank. If the valuation drops again, the effect reverses.

Stock analysts also use the word, though somewhat differently. They mean by it the value a stock ought to have according to their calculation. If the market price is below that, the stock is considered undervalued. This is an opinion, not an accounting rule, and the two meanings are frequently confused.

In one’s own portfolio, the principle is encountered daily. The displayed portfolio value is nothing other than a fair-value calculation using current prices. A typical mistake is to regard this value as certain. It only holds for the moment, and only as long as there are buyers at that price.

Subscribe free. Unsubscribe the second it sucks.

High-signal news across AI, business, UX, and tech. Every morning.