Für neue Autoren:
kostenlos, einfach und schnell
Für bereits registrierte Autoren
53 Seiten, Note: 5.75 von 6
2 Asset prices and their effects on the real economy
2.1 Asset prices and asset bubbles
2.2 Effects on the real economy
2.2.1 Wealth channel
2.2.2 Liquidity channel
2.3 Arguments pro and contra monetary action
3 Counterfactual analysis
3.1 Basic model
3.2 Asset prices model
3.5 Parameter estimates
3.6 Impulse response tests
4.1 Loss function
4.2 Baseline results compared with reaction to asset prices
4.3 Optimal policy parameters
4.4 Robustness tests
4.4.1 Different baseline error terms
4.4.2 Random shocks
4.4.3 Different magnitude of the wealth channel
Appendix A: Graphs and statistics
Appendix B: MATLAB code
List of figures and tables
Figure 1: United Kingdom output gap, inflation and short term interest rates 1994-2009
Figure 2: United Kingdom financial assets and real estate prices (logarithm)
Table 1: Model parameters
Figure 3: Unit shock to aggregate demand
Figure 4: Unit shock to aggregate supply
Figure 5: Unit shock to short term interest rate
Figure 6: Unit shock to financial assets fundamentals
Figure 7: Unit shock to real estate fundamentals
Table 2: Monetary policy parameters for counterfactual scenarios
Figure 8: Model results of scenario 'heavy reaction to RE assets only' and baseline
Figure 9: Deviations between scenario 'heavy reaction to RE assets only' and baseline
In modern economies it is the central bankers’ job to ensure the economic and financial stability needed to maximize the society’s welfare. As such, they need to take into account everything that is capable of significantly derailing the economy such as high inflation or a shock to domestic or foreign demand. If the central bank identifies such a risk, it reacts accordingly to minimize the adverse effects on the economy. Positive developments on asset markets however were often not taken into account beyond their impact on inflation and indication of output growth, even if asset market crashes that often follow and the financial turmoil they cause are by no means a new phenomenon (Reinhart & Rogoff, 2008). Only once markets began to fall and took the economy with them, central banks stepped in and tried to limit the damage. ‘Benign neglect’ and ‘mopping up afterwards’ were terms used to describe the approach. But in recent years criticism of this approach has been growing and especially after the near meltdown of the (developed) world’s economy in 2008 where ‘mopping up afterwards’ cost and still costs enormous sums, we have to reconsider if asset markets should be left alone or a more proactive monetary policy is warranted.
Diverse literature on the general subject of monetary policy regarding asset price ‘bubbles’ (the term ‘bubble’ is used as little as possible as it is imprecise; the term misalignments is used instead) including arguments both pro and contra. But while most try to answer the question whether or not monetary policy should react to asset prices in general, this thesis tries to ascertain if monetary policy should react differently to individual asset classes, as certain asset classes have a higher impact on households’ aggregate demand than others. Specifically, we look at financial assets and real estate assets as these two asset classes are both economically highly significant but have distinct differences in certain characteristics. It is assumed that real estate wealth is more evenly distributed between households than financial wealth and that price changes in real estate therefore should have a higher impact on consumer spending and thus aggregate demand.
The analysis is conducted with a dynamic structural general equilibrium (DSGE) model. The model is augmented with two wealth channels for real estate and financial assets and then exposed to a series of shocks to examine its behavior under both a baseline scenario reflecting reality and several counterfactual scenarios with altered monetary policy. The shocks are calculated on the basis of the economic data of the United Kingdom in the last 15 years to gener- ate a real example for our analysis. This thesis thus tries to answer the question if the central bank (in our case the Bank of England) should have reacted more to price changes in real estate than financial wealth, assuming it did not react in reality. The instrument under consideration is only the short-term interest rate as the main monetary tool. Other possible approaches to tackle the issue such as reserve requirements, regulatory approaches or quantitative easing as performed by the Bank of England (BoE) mainly in 2009 are therefore not looked at.
The remainder of this paper is structured as follows: In order to establish a solid foundation for further analysis, the next section deals with theoretical aspects of asset prices and their effects on the real economy. Also a short overview of the arguments pro and contra proactive monetary policy regarding asset prices put forward in the academic debate so far is given.
In the third section the macroeconomic model is developed and explained. After a detailed dissection of the model in section 3.1 and 3.2, the methodology underlying the counterfactual analysis is discussed in section 3.3. Details on the economic dataset, the parameters used and an analysis of the impulse response tests build the rest of this section.
The fourth section describes and analyses the results of the counterfactual scenarios. Beside the parameter sets used in the counterfactual scenarios, also the optimal policy parameters are determined. In addition, several robustness tests altering assumptions of the model are conducted to test the general conclusions.
The fifth section concludes and discusses the applicability of the findings to set future monetary policy.
Assets as a part of an individual’s, a firm’s or a country’s capital stock come in many forms and almost as many ways exist to define their value and price. Value and price are here distinctly separated as those do not have to be identical in times of market distortions. The price of an asset is formed in markets or if not available in individual trades while its (intrinsic) value is determined by economic fundamentals. But as the efficient market hypothesis (EMH) suggests that all available information on an asset is reflected in its price at any given time, its value given the information available should always be mirrored by the asset’s price in a world of efficient markets and no regulatory distortions (Bernanke & Gertler, 2000). So just until a few decades ago many were skeptic if persistent and destabilizing misalignments of asset prices even existed in a world of rational profit-maximizing agents (Gilchrist & Saito, 2006). In recent literature however there is a broad consensus on the existence of asset misalignments and asset bubbles. The strand of behavioral economics evolved with the insight that economic agents do not always act rationally (for a good introduction in the context of the dot-com bubble see Shiller (2005) or generally Akerlof/Shiller (2009) ). But also under the assumption of rationality one can argue in favor of the evidence of asset bubbles, e.g. in case of asymmetrical information (Diba & Grossman 1983, LeRoy 2004). This leads us to the definition of an asset bubble which is shared by most economists today and also used in this thesis: A bubble is a large and over a certain time period persistent (positive) misalignment of prices and fundamental values of an asset. We don’t define an exact threshold of how large the misalignment has to be to qualify as a bubble and do not exclude ‘negative bubbles’ i.e. prices far below fundamental values, although in practice most price developments identified as a bubble are positive. This definition therefore slightly differs from the common public view of a bubble also portrayed by Kindleberger (2009) as “an upward price movement over an extended range that then implodes.” (thus the term ‘bubble’).
In practice this definition is not easy to apply, as although we can readily evaluate the present price for most assets, the fundamental value is harder to define and even harder to measure. In general, the fundamental or intrinsic value of an asset is assumed to be the sum of its expected future dividends (Barlevy, 2007) and most models build on this assumption. Standard discounted cash flow (DCF) models sum up the discounted values of (expected future) dividends (for equity or debt) or rent income (for real estate, adjusted by real estate specific factors like population growth or interest rates). In the easiest case we have fixed and known cash flows (i.e. risk free debt) so the only determinant we have to form an expectation about is the discount factor. Mostly however, we also require some expectations about the size and probability of future cash flows (e.g. equities). For some other assets, e.g. currency or precious metals, defining the fundamental value is more difficult, as there are no dividends. For currency pairs some theoretical foundations like the interest parity or purchasing power parity can give us an idea of a ‘fair value’, but those concepts do not always withstand empirical evidence. If we even consider a single currency, i.e. cash as an asset, the value is purely expectational, the fundamental value zero: Money is only worth as much as we trust it to be worth tomorrow, the intrinsic value is zero in the case of electronic money or close to zero in the case of paper bills. One could therefore consider money the biggest asset bubble of all (The Townsend (1980) model of money outlines a good theoretical explanation of this argument). In the model used for the counterfactual analysis we use a standard DCF model by defining the fundamental value by expected future output (determining dividends) and expected future real interest rate (determining the discount factor).
These difficulties in determining the fundamental value make the identification of asset price misalignments very difficult, if not impossible. Even retrospectively, it is not always possible to reach a consensus on whether price development was misaligned or fundamentally justified. This is an argument often formulated against monetary action to contain asset bubbles. If the central bank cannot identify asset bubbles with relative certainty and, even harder, identify them with a reasonable lead time, reacting to asset price misalignments is at least useless if not harmful. But even critics of proactive monetary policy today agree that central banks always have to act under uncertainty even when measuring GDP and - due to knowledge of their own plans and unbiasedness - have in fact an advantage in recognizing asset price developments (Posen, 2006). Nonetheless, even with superior knowledge it might be politically unfeasible to contain asset price misalignments or just ‘lean against the wind’ if the public does not view prices as out of line with fundamentals, as it is not a popular action to raise interest rates and thereby slow down the economy. The central bank therefore needs very good arguments as to why it wants to hike rates, as people tend to dislike disturbances while getting rich.
In our model in the next section, the central bank is assumed to have full knowledge of the size of any asset price misalignments as well as of its linkage and magnitude to the real economy via the wealth channel.
If one excludes secondary effects of large asset price volatility, as e.g. effects on consumer price inflation (a good example is the spike in consumer price inflation in 2008 due to soaring oil prices also included in the sample period), then asset price misalignments and the resulting volatility in case of price corrections are per se of no concern to the central bank. Only when those developments have possible negative effects on the real economy (or more precisely: possible negative effects to the central bank’s goals), possible monetary action to prevent or contain them may be justified. The risks to the economy as a whole are primarily posed by the possibility of rapid price corrections (‘the bubble bursts’), not the bubble, i.e. price misalignments, per se. The larger the possible relative price correction (bubble size) and proportional effect on the real economy (bubble linkage) are, the larger the risk. Here we can identify different characteristics of the two types of asset bubbles (financial assets and real estate) examined in this thesis: While the size of financial asset bubbles is mostly large but its linkage is limited, real estate bubbles are smaller in size but typically highly linked with the real economy. If the identified risk in the form of an expected loss of the real economy due to a rapid price decrease is greater than the cost of monetary policy implied by raising interest rates to prevent or at least moderate it, the central bank should react to asset price misalignments. In the following we look at the two main channels via which asset prices have an effect on aggregate demand.
The wealth channel traditionally describes a positive linear relationship between asset prices (be it stocks or real estate) and household consumption, therefore increasing aggregate demand if asset prices rise (Mishkin, 2007). Some refer only to households when talking about the wealth channel, and separate an investment channel when referring to firms investing more if asset prices and especially the firm’s stock rise (also called Tobin’s Q). Here, the two are subsumed, as they describe the same relationship asset prices-aggregate demand, only for separate groups of actors which are not separated in our model. Our wealth channel thus describes the positive linear relationship between asset prices and aggregate demand; any nonlinear effects are subsumed as the liquidity channel treated below. The wealth channel is also part of most large scale macro models as the FRB/US used by the Federal Reserve (Reifschneider, Tetlow, & Williams, 1999) or the MCM of the European Central Bank (Willman & Estrada, 2002).
The basic idea of the wealth channel, a marginal propensity to consume out of wealth can be theoretically explained by the life-cycle theory first developed by Modigliani (Deaton, 2005) stating that individuals mainly put aside wealth to be able to consume it later (e.g. in retirement). Evidence of this marginal propensity to consume out of wealth can also be found in empirical data, with the magnitude being remarkably stable in the long run. The most important empirical finding however is significant differences in magnitude of wealth effects in different asset classes. Especially between financial assets and real estate the magnitude of the wealth effect seems to differ with evidence of the real estate effect being stronger, but the empirical data is not clear cut (Mishkin, 2007). For the counterfactual analysis however, we assume it to be stronger (in the baseline scenario), but general conclusions about the monetary reaction in the case of wealth channels of varying scale would remain valid even if the financial wealth effect would in fact be stronger.1
There are many theoretical arguments for both financial or real estate wealth having a stronger effect. The main argument for a stronger real estate effect is the fact that real estate wealth is far more evenly distributed across households and that the representative household has a larger fraction of its total wealth invested in real estate than financial wealth. This fact is valid for all OECD economies in the past 30 years (Catte, 2004) including our country of interest, the United Kingdom. Following this argument the stronger effect would not be due to different marginal propensity to consume but rather to the fact that real estate is far more common and interlinked with household’s consumption plans. The main counter-argument is the fact that real estate is not just an investment to accumulate wealth, but also renders a service in the form of housing. Therefore a rise in real estate prices may increase nominal demand but not real demand, as the representative household will not sell its now higher valued house. Also, it is not as easily possible for a household to sell a part of its real estate wealth as it is with financial wealth, but this is increasingly changing due to home equity loans which enable households to consume out of their real estate wealth without having to sell the entire house. Also, in general, real estate prices are less volatile than financial asset prices, especially stock prices. An increase in real estate prices may therefore be deemed more permanent than an increase in financial asset prices. For an extensive analysis of the different general arguments concerning the wealth effect refer to Mishkin (2001) or Bayoumi (Bayoumi & Edison, 2003). A good summary of arguments focusing on real estate are found in Zhu (2005) or Belsky and Prakken (2004). A study of the real estate wealth effect in the United Kingdom is found in Attanasio et al. (2004).
Additionally to a linear relationship, asset price developments can also have non-linear effects on aggregate demand, accelerating wealth effects in case of large price movements. A large price increase can cause a boom in consumption and investment, e.g. because higher asset values can collateralize more credit, thus multiplying the effect. A large price decrease or crash however can destabilize balance sheets of both firms and households and cause output to falter due to liquidity problems, as illiquid assets cannot be easily sold in times of rapidly decreasing value of a firm or a household’s assets. Also, banks will be less willing to give credit when the assets supposed to serve as collateral are decreasing in value or are expected to decrease. In extreme cases this can lead to a credit crunch as experienced in 2008. While all linkages and relationships in such a case are very complex, it seems clear that once prices fall and credit dries up, a flight to liquidity can start a vicious cycle, where prices fall even further and entire markets can get illiquid. Such a ‘financial accelerator’ is explained very well by Bernanke, Gertler and Gilchrist (1998), who implemented it in their model used to analyze monetary action towards asset bubbles (Bernanke & Gertler, 2000). A financial accelerator in the context of a DSGE model similar to the one used here can be found in Christensen (2005), one with a model of the banking system in von Peter (2004). In the model used here no financial accelerator is implemented for simplicity as we focus on the wealth channel.
Before we go on to the counterfactual analysis, let us quickly review the main arguments for and against a monetary reaction to asset price misalignments. First the arguments for a proactive policy are listed, followed by the arguments against it recommending a passive approach of ‘mopping up afterwards’. When referring to proactive monetary policy, we speak of a reaction to asset prices over and above their impact on inflation. The question if asset prices should be included in a proper measure of inflation and to what extent does not concern us here, as the question of the proper definition of inflation as a central bank’s target has to be answered in both proactive and passive monetary policy.
The first and most important argument in favor of a proactive policy, to which even critics agree, is the notion that it would be desirable to limit the fallout of bursting asset price bubbles to the real economy if feasible. While this seems trivial, it implies the question is only about feasibility, not desirability. Also, as Filardo (2004) points out, as central banks do react to economic downturns following asset price busts, they should out of symmetry also react to price buildups. An asymmetric reaction is inconsistent, as there should either be a symmetric reaction to asset prices in the case of proactive policy or none at all if the central bank chooses a passive stance (Roubini, 2006). In addition, reacting only to asset price downturns could result in a moral hazard problem as market participants would rely on the central bank to ‘save’ them if prices fall, thus eliminating the downside risk. This moral hazard was also named the ‘Greenspan put’ after the Chairman of the Federal Reserve at the time of the dotcom bubble (Miller, Weller, & Zhang, 2002).
Most analyses using macroeconomic models point to optimal policy rule parameters being non-zero regarding asset price misalignments. This does not mean central banks should prick asset price bubbles, but that there should be at least some reaction ‘leaning against the wind’ as Cecchetti (2000) first pointed out. Also as Robinson (2005) shows, this reaction also helps to avoid the problem of the zero lower bound, as central banks cannot lower the interest rate below zero even if optimal policy rules would advocate it. Other analyses of optimal policy rules coming to the same result are found in Kent and Lowe (1997), Filardo (2004), Kontoni- kas (2005) and Bask (2009).
When looking at the arguments against proactive policy again the desirability of containing asset price swings is not disputed, but rather the possibility of practical implementation. First of all, for a central bank to successfully prevent asset misalignments the central bank has to be able to detect and identify them reliably. And not only does the central bank have to identify the misalignment for itself, but it also needs a case strong enough to explain to the public (and the markets) why it sees the necessity to raise interest rates. However, if there was general agreement that there is a misalignment, the misalignment itself would not exist. This identification problem where one has to determine if a price increase is fundamentally justified or in fact misaligned was also one of the arguments used by Bernanke/Gertler, the most prominent critics of proactive policy, to argue for strictly passive monetary policy (Bernanke & Gertler, 2000). They find a policy rule indifferent in asset prices to optimize welfare in their model (Bernanke & Gertler, 2001), however their results were challenged by Cecchetti (2000).
The main argument against proactive policy is directed against the tool at hand, short-term interest rates. Following the argument, interest rates are ‘too blunt a tool’ (Posen, 2006). If there is money to be made in a bubble and prices rise significantly, investors will not alter their decisions when interest rates are raised by a few percentage points. On the other end however, the same interest rate hike by a few percentage points will cause enormous costs to society by considerably slowing down the economy. Maybe even the interest hike itself will cause the economic downturn the central bank wanted to prevent. In a multi-country VAR study Assenmacher and Gerlach (2008) show that the impact of a monetary policy hike on property prices is only about three times the impact on GDP. So to prevent a typical house price misalignment where prices increase by 20%, the interest rate hike needed would depress GDP by more than 6%, which is even worse than the severe recession in most industrial nations in 2009.
Finally, the central bank does not have perfect control over interest rates, as it is not the only source of liquidity in modern financial markets. As we have seen very clearly during the credit crunch in 2008, in times of extraordinary market sentiment, be it over-confidence or panic, the actual interest rates available to market participants can be very different from the central bank’s set rates. The central bank’s monetary policy acts with considerable lag, which further complicates a possible proactive policy. The central bank has to be certain about the present stage of an asset misalignment, as a proactive reaction (i.e. raising interest rates) could prove disastrous if the bubble is about to burst on its own, meaning the interest hike comes into effect during the economic downturn. Posen (2003) shows this in the case of Japan’s real estate bubble and following ‘lost decade’.
The model used in this thesis is a structural rational expectations model implementing the standard three macro sectors aggregate demand, aggregate supply and monetary policy. It is basically an augmented version of the model specified by Ball (1997) and Svensson (1997) and follows the work of Kontonikas (2005) by adding and modeling asset prices to influence aggregate demand as well as monetary policy (in the counterfactual scenario). For simplicity the micro-foundations of the model are not reviewed in this thesis, but can be found in McCallum and Nelson (1998) among others. The three macro sectors are given by the following equations:
Abbildung in dieser Leseprobe nicht enthalten
where [Abbildung in dieser Leseprobe nicht enthalten] is the output gap, nt is the inflation rate, it is the one-period interest rate (i.e. 3- months-LIBOR) as the monetary policy instrument, [Abbildung in dieser Leseprobe nicht enthalten] is the natural rate output implying [Abbildung in dieser Leseprobe nicht enthalten] = [Abbildung in dieser Leseprobe nicht enthalten] — [Abbildung in dieser Leseprobe nicht enthalten] with [Abbildung in dieser Leseprobe nicht enthalten] being the actual output value, r is the equilibrium real interest rate, n* FA RE is the equilibrium inflation rate (i.e. the inflation target by the central bank) and qt and qt are the (log) financial assets and real estate prices respectively ( * indicating the fundamental values). nt, £t and dt represent exogenous shocks to output, inflation and the nominal interest rate and are later taken from the actual historical data to perform the counterfactual experiment.
Equation (1) models aggregate demand as an optimizing IS-equation where current output depends positively on expected future output and asset prices, and negatively on the real interest rate which is formed by the past period’s nominal interest rate it and the current period’s inflation П. To enhance realism one could also use Et_i[ П ] as investment decisions are made by taking the expected future inflation into account when looking at the nominal interest rate. The differences in the model’s behavior are quite small however, so the current inflation rate is used to simplify the calculation of the data error terms for the counterfactual analysis. The impact of expected future output on current spending is theoretically justified by the thought that households take their expected future income into account when deciding about current consumption. More technically, the term can be derived from the first order Euler conditions for the household’s optimal consumption choice problem, which is shown in the context of an optimizing general equilibrium model by McCallum and Nelson (1999).
The third and fourth term of the equation signifies the wealth channel, where aggregate demand depends positively on past asset prices via consumption effects by households and investment balance sheet effects by firms. Rising asset prices increase household wealth, which in turn increases consumption and aggregate demand. Also rising asset prices increase firm’s valuations which in turn improve their ability to borrow and invest (see Bask (2009) or Smets (1997)). Falling asset prices have the opposite effect. While there is empirical evidence that rapidly falling asset prices (a ‘crash’) can also have non-linear effects via a ‘financial accelerator’ (Bernanke, et al., 1998), the model does not incorporate such a financial accelerator for simplicity.
Equation (2) is a hybrid Phillips curve to model the price adjustment relation where inflation is determined by both expected future inflation as well as the output gap. By taking into account past inflation, hybrid Phillips curves model inertia in price formation processes (price stickiness) and thereby reduce the inconsistencies between purely forward-looking models used in earlier years and inflation data in reality. For a detailed explanation see Clarida, Gali and Gertler (1999). The output gap term passes on demand pressure on the inflation rate in the case of an overheated economy, which can also be justified empirically.
Equation (3) represents the monetary policy reaction function determining the central bank’s behavior. Depending on the coefficients y1 to y4, the central bank reacts with different magnitude to inflation, the output gap, financial assets and real estate prices. If y3, y4 = 0, the reaction function reduces to a standard Taylor rule (Taylor, 1993). The parameter y5 on the past interest rate depicts the central bank’s tendency to smooth the path of the policy rate to enhance predictability and stability of monetary conditions. The parameters y3 and y4 are later used for the different counterfactual scenarios with all other parameters held constant. The fundamental valuation of asset prices is taken from the asset prices model (see below) to determine - analogously to the output gap - asset price misalignments to which the central bank reacts. This of course implies the assumption that all price differences between actual and fundamental asset prices are misalignments which should be corrected, which is already a strong assumption. Furthermore as Kontonikas (2006) already points out in his paper, equa- tions (1) and (3) lead to the assumption that the central bank is fully aware of the impact of both financial assets and real estate prices on aggregate demand and its magnitude.
To perform the analysis, a model of asset prices and their fundamentals is needed as well to look at the effects of different economic outcomes on asset prices. The model is very simplified and consists of two equations for each asset class, one determining the current asset price level and one defining the current fundamental value. The equations are the following:
Abbildung in dieser Leseprobe nicht enthalten
The Equations (4) and (6) model the price development by reacting positively to past price increases and negatively to deviations from the fundamentals. This corresponds to a momentum trading trend and a mean reversion trend where prices return to their fundamentals. Following Kontonikas (2005) and Bask (2009) the model does not incorporate a deterministic probability of an asset price ‘bubble’ but rather allows prices to deviate from fundamentals without defining an asset price ‘bubble’ (for a deterministic approach see Berger & Kissmer (2006)). By lowering the coefficient b2 or c2 we reduce co-movement between actual prices and fundamentals and therefore get more price misalignments. Price movements by momentum trading allows for herd behavior commonly seen on asset markets and puts less restriction on the model than e.g. the approach of Bask (2009) who assumes technical trading (i.e. trading by technical analysis of price charts) and least squares learning. An example of a detailed model of momentum trading as well as empirical evidence of it is provided in Hong and Stein (1997), who provide just one of many counterexamples to Friedman’s famous claim that destabilizing speculators cannot exist due to natural selection.
Equations (5) and (7) represent the fundamental asset price valuation as a standard dividend model of asset pricing. The fundamental value depends positively on expected future output and negatively on the real interest rate formulated by the nominal interest rate and expected future inflation. This corresponds to expected future dividends (which we assume to depend on expected future output) and the discounting factor (the real interest rate) in the standard
1 For the numerical estimates of the wealth channels see the parameter estimates section.
Bachelorarbeit, 34 Seiten
Masterarbeit, 88 Seiten
Bachelorarbeit, 33 Seiten
Diplomarbeit, 105 Seiten
Seminararbeit, 13 Seiten
Masterarbeit, 69 Seiten
Masterarbeit, 59 Seiten
Masterarbeit, 63 Seiten
Der GRIN Verlag hat sich seit 1998 auf die Veröffentlichung akademischer eBooks und Bücher spezialisiert. Der GRIN Verlag steht damit als erstes Unternehmen für User Generated Quality Content. Die Verlagsseiten GRIN.com, Hausarbeiten.de und Diplomarbeiten24 bieten für Hochschullehrer, Absolventen und Studenten die ideale Plattform, wissenschaftliche Texte wie Hausarbeiten, Referate, Bachelorarbeiten, Masterarbeiten, Diplomarbeiten, Dissertationen und wissenschaftliche Aufsätze einem breiten Publikum zu präsentieren.
Kostenfreie Veröffentlichung: Hausarbeit, Bachelorarbeit, Diplomarbeit, Dissertation, Masterarbeit, Interpretation oder Referat jetzt veröffentlichen!