**2. Literature review**

to technological advancements across all industries. Not only more people connect to the internet every day, also devices actively communicate with one another, coined under the term "the internet of things". Moreover, recent initial public offerings of well-known internet companies such as Groupon, LinkedIn, Facebook, Alibaba Group Holding Limited and funding support programs in the United States, the European Union and China drive the interest

When looking at the development of certain equity indices in the United States, Germany and China, the rates show record or close to record peaks as seen on **Figure 1**. For the Nasdaq Composite Index such rates were last seen during the dot-com bubble while the Deutscher Aktien Index 30 is almost twice as high as during the dot-com bubble and before the recent financial crisis of 2008. For the China Securities Index 300, the sources from the Bloomberg database even state price-to-earnings ratios as high as 220 times reported

Bringing these circumstances into a global context, an artificial economic boost was created after the crisis that made equity indices soar—just like before the dot-com bubble. Hence, the question arises how the situation as of December 2015 is different if it is at all. Are there signs

The dot-com bubble is one of the most disputed bubbles that occurred in the last decades. Each bubble can be modelled according to some rules. Welfare analysis with empirical prediction is subject to examine [3]. Also, the pure statistical tests are suitable to describe bubble [4]. But alternative assessment can be provided by another angle of view on bubbles [5].

of investors for lucrative opportunities.

94 Financial Management from an Emerging Market Perspective

of an asset price bubble as some research argues [1, 2]?

Moreover, market bubbles are related to market volatility in general [6].

**Figure 1.** Increase of indices' values across regions. Source: Own elaboration by the authors.

profits.

Since the current situation of peaking indices is too new to expect extensive research on this topic, there are currently only a few similar articles available to the knowledge of the author [7, 8]. Their research, however, is either limited to a single stock, market or pursues another approach. Furthermore, the eventual assessment of a bubble can only happen in retrospect while today's perspective can only try to catch the market sentiment and interpret some key economic indicators.

#### **2.1. Asset price bubble**

By the definition of the majority of literature, an asset price bubble is characterised by a substantial deviation from the fundamental value of the asset [9]. On the other hand, according to the efficient market hypothesis, there are three levels of efficiency: weak, semi-strong, strong—which represent the incorporation of information into the price of securities [10]. Hence, under certain assumptions such as that investors perceive information uniformly, securities should trade at their fair value and deviating from the fundamental value should not be possible with respect to the level of efficiency. Despite the critique and controversy of the efficient market hypothesis in recent decades it was confirmed again in 2009 [11]. Nevertheless, severe price deviations did occur and cannot be explained by the efficient market hypothesis. Hence these deviations represent a key flaw in this hypothesis on which much of economy theory relies. In theory, rational investors are expected to base their valuations on fundamentals such as the intrinsic value of an asset which can be approximated by the discounted sum of future cash flows [12]. Since nobody can predict the future, however, projections of future cash flows are likely to be highly subjective. To tackle this issue, the assessment of a bubble can be mainly approached from two dimensions: magnitude and time.

While it is recognised that measuring the fundamental value at the very moment of the occurrence is hardly possible [13]. Therefore, it is suggested that a bubble should be measured by deviations of expected and realised returns over a defined period of time—even decades [14]. Obviously, this can only happen in retrospect since returns cannot be predicted with certainty, especially during times of turmoil in the markets [15].

Beyond the pure stock price, another crucial factor according to equity asset valuation is often taken into account: the price-to-earnings ratio or earnings in general. Commonly speaking, the higher the price-to-earnings ratio increases without substantial news to support a higher valuation, the harder it becomes to justify prices if there is little evidence for future returns other than the pure belief—for instance speculation.

Differentiating a plausible market development and an irrational bubble remains the subject of collective research on expectations and actual future cash flows. In retrospect, only the burst of a bubble, indicated by a substantial depreciation of prices, gives reason to assume that there was one in the first place. That is because a permanent change would require a fundamental reason and not constitute a bubble due to the lack of deviation. Also, there can be an overlap with different types of business cycles. This issue relates to the second dimension by which a bubble has to be assessed – time.

For example, there is an argument that the actual dot-com bubble was quite short and only lasted from 1998 to 2000 [16]. Even though, this might be the time of the actual substantial price deviation, it is important to take into account the circumstance which led to it in the first place. Therefore, a five-stage approach that was described and later refined is favoured in this paper [17]. This is supported by another research, which used a three-stage to five-stage approach, too [18].

Since a single value or theory is unlikely to explain the workings of a complex event of an asset price bubble which is the interplay of irrational humans and algorithm driven economics, it is more likely the combination of these factors.

#### **2.2. Explanations for the occurrence of asset price bubbles**

Even though there currently is no guaranteed empirical method to detect an asset price bubble [19], especially in real time, there are still again mainly two methods to approach this issue [20]. The first one is based on rational expectations of the efficient market hypothesis, and the second one on behavioural finance.

Firstly, there are empirical and technical models that are based on the efficient market hypothesis [21, 22]. Even though, such models are able to provide sufficient explanatory power, they are for example limited to historic data [23], certain stocks or companies [24] or more generally, to certain assumptions which were true in the past. Most importantly they lack the major factor of irrational human behaviour that can be observed in certain situations at the stock market, especially during uncertainty due to politics or the introduction of an innovation. Under the category of rational expectations also fall the established economic theories that take into account monetary policies, investment and consumption cycles. For example, there is also an evidence for connecting the emergence of bubbles with the credit creation policies [25].

Secondly, there are qualitative metrics which take into account soft factors such as the current sentiment of the market and irrational human behaviour, including biases as coined under the term behavioural finance. Psychological factors, especially during turbulent times, gain importance while traditional measures lose in relevance [26]. Concepts like herding behaviour and greater-fool theory trump rational economic assumptions such as the efficient market hypothesis and investors willingly forego arbitrage opportunities. Also, soft factors such as human capital, strategic alliances, joint ventures and internet popularity gain in importance [27–29]. Since only forecasts about future cash flows can be made, expectations play an integral role to the development of the market. For example, there is a study showing that during the dot-com bubble financial analysts were more optimistic about internet stocks [30]. Especially in such circumstances fundamental values start to become irrelevant to investors [31]. Also, it must be taken into account that even though there might be an asset price bubble, it can be necessary due to circumstances to ride that bubble instead of acting rationally against the majority of market participants [32]. After all there is nothing so dangerous as the pursuit of a rational investment policy in an irrational world. The financial sector plays also a significant role here [33, 34]. A good example is technical analysis with the Federal Reserve System model which is extensively criticised by theorists for their flaws but used by practitioners nevertheless. The sheer beliefs and usage of such concepts substantially influence the market even without sound academic background [35]. Furthermore, restrictions like short selling, set by a higher body or lock-in periods for stock options, may skew and alter the economic normalisation processes [36]. Moreover, history shows that politicians are reluctant to take countermeasures before an imminent economic threat can be proven because acting on a false alarm would result in a loss in economic output which in consequence would have negative transitioning effects throughout society [37, 38].

So, taking into account the irrational human behaviour is an important factor. Therefore, a holistic combination of both theories efficient market hypothesis and behavioural finance is expected to yield a better result for the explanation and assessment of a bubble formation.

#### **2.3. Other effects**

For example, there is an argument that the actual dot-com bubble was quite short and only lasted from 1998 to 2000 [16]. Even though, this might be the time of the actual substantial price deviation, it is important to take into account the circumstance which led to it in the first place. Therefore, a five-stage approach that was described and later refined is favoured in this paper [17]. This is

Since a single value or theory is unlikely to explain the workings of a complex event of an asset price bubble which is the interplay of irrational humans and algorithm driven econom-

Even though there currently is no guaranteed empirical method to detect an asset price bubble [19], especially in real time, there are still again mainly two methods to approach this issue [20]. The first one is based on rational expectations of the efficient market hypothesis, and the

Firstly, there are empirical and technical models that are based on the efficient market hypothesis [21, 22]. Even though, such models are able to provide sufficient explanatory power, they are for example limited to historic data [23], certain stocks or companies [24] or more generally, to certain assumptions which were true in the past. Most importantly they lack the major factor of irrational human behaviour that can be observed in certain situations at the stock market, especially during uncertainty due to politics or the introduction of an innovation. Under the category of rational expectations also fall the established economic theories that take into account monetary policies, investment and consumption cycles. For example, there is also an

evidence for connecting the emergence of bubbles with the credit creation policies [25].

Secondly, there are qualitative metrics which take into account soft factors such as the current sentiment of the market and irrational human behaviour, including biases as coined under the term behavioural finance. Psychological factors, especially during turbulent times, gain importance while traditional measures lose in relevance [26]. Concepts like herding behaviour and greater-fool theory trump rational economic assumptions such as the efficient market hypothesis and investors willingly forego arbitrage opportunities. Also, soft factors such as human capital, strategic alliances, joint ventures and internet popularity gain in importance [27–29]. Since only forecasts about future cash flows can be made, expectations play an integral role to the development of the market. For example, there is a study showing that during the dot-com bubble financial analysts were more optimistic about internet stocks [30]. Especially in such circumstances fundamental values start to become irrelevant to investors [31]. Also, it must be taken into account that even though there might be an asset price bubble, it can be necessary due to circumstances to ride that bubble instead of acting rationally against the majority of market participants [32]. After all there is nothing so dangerous as the pursuit of a rational investment policy in an irrational world. The financial sector plays also a significant role here [33, 34]. A good example is technical analysis with the Federal Reserve System model which is extensively criticised by theorists for their flaws but used by practitioners nevertheless. The sheer beliefs and usage of such concepts substantially influence the market

supported by another research, which used a three-stage to five-stage approach, too [18].

ics, it is more likely the combination of these factors.

96 Financial Management from an Emerging Market Perspective

second one on behavioural finance.

**2.2. Explanations for the occurrence of asset price bubbles**

There are argues that there may be many reasons to cause asset price bubbles [39]. There are several research papers which take also nonfinancial but rather information technology oriented factors into account that might influence investors' decisions when buying technology stocks. The reasoning behind such measures is the Anglo-Saxon approach which focuses on growing big and gaining market share at high costs in the beginning while becoming profitable later [40]. Such factors include, for example, page views or visitors and sales in contrast to traditional measures such as earnings before interest, taxes, depreciation and amortisation or net income [41, 42].
