Term Structure as a Real World Macro Predictor: US and Beyond

5133 words (21 pages) Business Assignment

12th Jun 2020 Business Assignment Reference this

Tags: Business AssignmentsEconomics

Disclaimer: This work has been submitted by a student. This is not an example of the work produced by our Business Assignment Writing Service. You can view samples of our professional work here.

Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of BusinessTeacher.org.

Arturo Estrella and Gikas Hardouvelis introduce “The Term Structure as a Predictor of Real Economic Activity.” Knowing that the term structure suggests a timeline or fixed period, the goal is to continue assessing the macroeconomic environment in the United States using term structure as a predictive measure. A positive slop is symbolic of future increases in real economic activity, which will be tested throughout this analysis. STATA is used for data output, where the data is analyzed under the premise established by Estrella and Hardouvelis. A positive slope of the yield curve shows enhanced predictive power over the index of leading indicators. This analysis shall test the positive slope of the yield curve and its predictive power, where the data is focused on the current administration (President of the United States), prior administration, and prior economic trends in the United States.

Information in the slope is not independent of monetary policy. Estrella and Hardouvelis look at leading indicators, real short-term interest rates, and connected variables such as lagged growth, lagged rates of inflation, and long-term interest rates. The study by Estrella and Hardouvelis examines whether the positive slope of the yield curve is going to be as strong of a predictor, comparatively, when recognizing these other variables. The predictive power of the positive slope is explored throughout this assessment, where the hypothesis appreciates its value as being as powerful as the other factors, or inter alia. Term Structure is proven an accurate predictor of economic trends.

Term Structure

Prior to analyzing the data itself, a strong understanding of the term structure is necessary. Put simply, the term structure comes with expectations (Gurkaynak and Wright). When considering the term structure of interest rates, the topic is going to have much importance to investors, and also to policymakers, who wish to extra macroeconomic expectations from the longer-term interest rates (Gurkaynak and Wright). Here, action will be taken to influence those rates, where interest rates are one of the prime examples of the term structure and its use.

Under the term structure, one has expectations (hypothesis) of what will be paid, when the term is over. In its simplest form, a creditor lends $5 to a debtor. The debtor has a week to pay back the $5 at an interest rate of 10%. At the end of the week, the term structure hypothesizes that the debtor will pay $5.50. Once paid, this scenario is closed and the term structure has met its expectations. Of course, market segmentation is also necessary to understand in order to appreciate the term structure of interest rates during particular financial events, such as a crisis (Gurkaynak and Wright).

Understanding the scope of term structure is significant in trying to make sense of macroeconomic development in the United States. Real economic activity comes with both predictable variables and those non-predictable, where external forces are going to disrupt the macroeconomic environment (Gurkaynak and Wright). Thus, knowing the parameters of the term structure is important from a literal sense. Yet, theory would suggest that the term structure on its own cannot be considered an accurate predictor of economic activity, in the United States or throughout the global environment. In continuing to explore Estrella and Hardouvelis, these trends must be considered and the idea becomes looking at these outliers or external forces that would disrupt what is an otherwise favorable predictor of economic activity.

Economic Policy Uncertainty, Market Returns and Expected Predictability

Economic policy uncertainty (EPU) is an index recognized by Frederick Adjei and Mavis Adjei. The impact of the level of EPU on the conditional mean of market returns is assessed, where the research examines the predictive power of EPU on future market returns (Adjei and Adjei). Even after controlling for business cycle effects, EPU is inversely related to contemporaneous market returns (Adjei and Adjei). This is influential in our assessment of the predictive power of real economic activity under the term structure. The term structure is applied in the research, as demonstrated by Adjei and Adjei, as well as Estrella and Hardouvelis. Adjei and Adjei recognize this as being the first study examining the predictive power of EPU on future market returns. We apply these findings to the data discovered under STATA, where our analysis takes a long-term approach spanning the periods recognized. Specifically, the data looks at the terms prior to PO, the terms of PO, and the current term (PT), up until June, 2019. With applying Adjei and Adjei’s findings relative to predictive power of EPU, there is consistency in appreciating Estrella and Hardouvelis.

GRAPH 1 is a predictor of real economic activity, under the slope recognized by Estrella and Hardouvelis. Here, the slope begins during the Recession. It is important to note the impact of the Recession, and its end point. 2019 is the current term (PT), where 2010-2015 is recognized as the prior term (PO). With this graph, we are able to analyze how the predictive power remains consistent, which is data-backed evidence to support the claims being made by Adjei and Adjei, as well as confirming the hypothesis of Estrella and Hardouvelis. Thus, this trend will be continued and explored with respect to the data output (Appendix) as well as the literature review moving forward.

Real Economic Activity

A positive slope of the yield curve represents a future increase in real economic activity (Estrella and Hardouvelis). Here, real economic activity is recognized as consumption, consumer durables, and investment (Estrella and Hardouvelis). The positive slope has extra predictive power over the index of leading indicators, real short-term interest rates, lagged growth (in economic activity), and lagged inflation rates (Estrella and Hardouvelis). So, the authors are promoting a positive slope of the yield curve as being a sound indicator of a future increase in real economic activity, and promote this trend to begin their analysis of real world predictors (in the United States, but also applicable elsewhere) (Estrella and Hardouvelis).

The positive slope, historically, promotes inter alia. According to the analysis, the slope is suggestive and predictive of factors independent of monetary policy (Estrella and Hardouvelis). Just as inside information would be useful to a trader or investor, but is often unrecognized by those outside of the market (but still in the macro environment), the research has considered the slope to be tertiary information. Whereas it is useful when making a private investment or setting a policy, the typical scope of macroeconomics is not going to follow this slope and the positive pattern (Estrella and Hardouvelis). The premise here is that the slope is being underutilized as a predictor of macro activity in the wider environment. The slope is a strong predictor, where the positive slope is going to show an increase in real economic activity, not only for investors and insiders, but for the economy as a whole. This theory will be tested with respect to the United States and current trends, expanding this theory and hypothesis beyond 1991, when the study was conducted by Estrella and Hardouvelis.

GRAPH 2 uses the same trends in analyzing the data and making sense of the predictive power of the slope. Here, Estrella and Hardouvelis have introduced the positive slope, which begins at the end of the recession and the initial term (PO). Then, the slope shifts into the current term (PT), as is recognized by the data output (Appendix) and all appropriate graphs. As discovered, the PO Trend and PT Trend show similar returns under the predictive power, further confirmation that the predictive power of the slope is efficient, and continued analysis that the term structure is sound in analyzing predictive measures moving forward. Moreover, the trend suggests that predictive power is not limited by a term, where the PO Trend shows similarities to the PT trend, despite external changes to the economic environment. While the recession has ended prior to the PO Trend identified in Graph 2, other factors have caused the shift. Even so, there is a general consensus based on 95% confidence interval that predictive measures at the macroeconomic level are going to hold constant. Moving past the 2019 (Current) Trend, where PT is still operating as a trend, we expect that these predictive measures are going to continue not only in this term, but in the next term.

Data Collection

In order to assess the predictive status of a positive slope, there is a need to collect data. Specific data is pertinent in understanding the predictive nature of the economy, and whether there is validity in making a claim that it is time-sensitive, or that a term structure is a valid predictor of economic activity moving forward. Examples of data include real GDP; unemployment; labor participation rates; real wages; debt to GDP; Consumer Price Index; and the Dow Jones Industrial Average. For each of these economic indicators, it becomes critical to test the growth rates in recent years. Understanding that there has been what is considered a recession to the macro environment, the time series predictor or the structure used to make predictions would have accounted for all of these factors (collected data) as they are not outliers, but sound components of economic activity (DeJong and Chetan).

Economic Growth Measures

It must be understood that any economic growth measure is going to paint an essential picture of the economy moving forward (DeJong and Chetan). In predicting economic means in the United States using a term structure, one of the more feasible and logical ways to do this is to make a comparison between economic trends under current president Donald Trump, and former president Barack Obama. This is done with a hypothesis that Trump will continue an eight-year term of the presidency, as did Obama during his administration. So, the goal is to appreciate any data-driven evidence that the trends in economic growth under current president Trump (“PT”) are significantly difference when compared with the trends in economic growth (or recession) under the administration of Obama (“PO”). Of course, there is a need to also account for the inherited measures that PT would take on, which could have been influenced by PO.

The time series predictor and the term structure is a measure of growth, but must account for inherited value. The data finds that there is inherited value (positive or negative) that must be accounted for in order to use this as an indicator of growth (Estrella and Hardouvelis). Put simply, if the country (United States) is facing a deep recession, high unemployment rates, and catastrophic elements to the micro-based economy, this is going to put the new president (PT) in a vulnerable position. Alternatively, if the economy shows sound value and low unemployment rates under PO, then PT will have a significantly easier and more fluid position with respect to the economy. It must be understood that there are increasingly-valuable elements of the term structure that cannot be accounted for in a predictor (Estrella and Hardouvelis). The goal is to continue assessing whether this is a predictive measure of the economy, and what the data is going to show.

Keys to Analysis

As has been discovered, there will be two periods of focus in using this predictive tool as an indicator of economic success or failure in the United States. PT has inherited the economic parameters of PO. Whereas real world data may be incorporated, the figures themselves are only a reflection of the data, and their views or values are not being recognized as influential to the data itself. Rather, the goal is to align the current trends in the economy with the position of these trends, and how the structure is either going to be an accurate or inaccurate display of the data for the macroeconomic environment. We do account for the real world economic principle where PO left the economy and its current values to PT. This puts PT in a vulnerable position if the macroeconomic environment is in a prime position, but a poor and threatened situation if the economy is in shambles (DeJong and Chetan). This is simple economics, and does not need further explanation as it becomes clear it is easier to continue running a sound economy, as compared with having an unsound economy and trying to overturn past faults (DeJong and Chetan).

The data uses predictive measures in assessing whether PT is different than PO. More specifically, predictive data will be used in determining current trends to the macro environment, and determining whether economic growth under PT is any different than it was under PO. The findings will shoe that there is a continuation of the trends in economic growth that have been reaffirmed by PO, which would argue that the predictor (term structure) not only continues to prove worthy, but is recognized as being efficient in the current environment.

Trends and Predictive Measures

For each economic measure, we will plot the data for that measure. The end of the recession has been highlighted in GRAPH 1, GRAPH 2, and GRAPH 3. This is measured with the most recent available value, given an interpretation of the data. Then, a trendline is introduced depicting the economic state and aligned variables pre-PT (during PO and prior). The trendline is recognized as the average growth rate for that economic measure observed during PO. Again, this begins with the end of the “Recession” and continues throughout the period under PO. Moreover, ease of interpretation in analyzing the data is necessary. Thus, a vertical line has been introduced to indicate when PT began taking economic initiative, and PO would stop (end of term). The presidency is an accurate predictor of term structure because of the strict timeline, where we are able to recognize the transition from the recession to PO to PT, knowing that other factors are consistent.

A 95% Confidence Interval (CI) has been implemented for the trendline. Here, the 95% C reflects the range of values for the growth measure that may be expected to observe when there is no change in economic growth trends (DeJong and Chetan). That is, this is indicative of fluidity and consistency in the trends as a predictor based on the term and the structure of PO as compared with PT. We expect that the structure itself is going to be constant, and know that the time series is constant, where we are able to accurately (at 95%) predict whether the time series and the predictor are going to be an accurate depiction of the economy moving forward. The current year (2019) comes with change variables on a month-by-month basis, where the data current concludes with June, 2019 as a stoppage point, but can account for other periods throughout the year, given limitations.

If the economic growth measured under PT were to fall within the 95% confidence interval, we would conclude that there is no statistical evidence of a change in economic trends. Thus, prior to examining a reflective picture of what has been discovered in comparing PO to PT, we must paint a picture of what would be expected if the economy were to undergo a structural change, or a break in growth rates. In doing so, we use Graph 1 to plot the growth rate of real GDP from the official beginning of the Recession (2007) to the most recent time period (2019, June). GDP growth rates will be measured at the quarterly level (DeJong and Chetan). Here, we include a trendline (GDP growth rates) depicting this trend only when the Recession was impactful to the economy, which is between December 2007 and June 2009 (DeJong and Chetan). Here, the 95% CI is incorporated, where we are able to have (with 95 percent confidence) a measure of expected GDP growth considering the Recession would have continued, where the vertical line (Graph 1, Graph 2, Graph 3) is official in all depicted trends (see APPENDIX).

Real Macro Example

In order to provide housing for the citizens of the United States, or any country there is going to be a slower rate of growth when assessing the GDP. Of course, one can understand that such reforms are only going to slow the GDP growth on a temporary basis. Allowing individuals to live in homes that they can afford and empowering small businesses to begin producing and developing throughout U.S., will have long-term positive effects that will inevitably cause the economy to become healthier and the GDP to rise in the future (Walton). Similar trends have been recognized with respect to China, taking a real world approach similar to that in the United States (Orlik).

If individuals are unable to afford housing, their lifestyle becomes detrimental. These individuals are not going to buy and spend within the economy because of the external factors that disallow them from doing so. Furthermore, the small businesses that are unable to secure funds in order to get started are not going to produce goods or deliver services. While this may cause a temporary slowdown in terms of GDP growth, the long-term ramifications are beneficial as a healthier economy will be the result under the assumption that those who have secured housing will eventually buy and spend, and that those who are now able to start-up long-term businesses are able to sell goods which causes a circulation of funds within the economy, which is the case with China and further strengthens the hypothesis of the effectiveness of predictive (time series) measures (Orlik).

A Similar Picture

When repeating the exercise recognized in the APPENDIX (Graph 1, Graph 2, Graph 3), a similar pattern emerges. That is, all are going to paint a seemingly similar picture. There is no data-driven evidence suggesting the trends in economic growth under PT are going to be any different than the trends recognized under PO. Political outcries aside, the data is more suggestive of a predictive trends in the macro environment, where elemental data is more sufficient than any claims that the economy is in a stronger position under PT, as compared with the economy under PO. The recession has been considered as an effective measure in shaping the economy, in both capacities. We find that the recession shows stability in terms of how it is impactful toward the data for both PO and PT. That is, there is not favoritism given to the economic structure of the economy under PO simply because the recession was closer. Moreover, we account for the recession that has been inherited under PT, and the data still shows that there is limited, if any disparity concerning the economic trends and how they are impactful in the macroeconomic environment.

Continuance of a trend under similar guidelines is optimal in applying a real world economic approach. Mark Taylor finds predictive trends that must abandon the desire to look at structural changes, and those brought on by certain demographics. Our findings are going to follow this trend, and not incorporate smaller elements that would disrupt this real world approach (Taylor). For example, focusing on particular segments of the population and how they are not gaining employment at a productive rate would shift the predictive measures. We cannot predict particular changes in this capacity, where Taylor argues that these parameters should be removed from the wider scope, and ignored with the real world approach to the situational analysis. Here, Taylor’s findings have been incorporated into our findings with respect to the graphs that are highlighted in the Appendix. While one will recognize there are outlying trends incorporated, such as the unemployment rates among the particular groups, our findings uphold the term structure practice, and follow our hypothesis in proving this to be an effective element of predictive status (Taylor). PT has inherited a position under PO, but this does not cause a dramatic shift in our term structure and how it will view the macro environment.

GRAPH 3 accounts for the 95% Confidence Interval, integrating Unemployment Rate as a percentage change from the preceding point. As is the same with Graph 2, the trend begins with the End of Recession as the period. PO Trends are compared with PT Trends. As has been discovered throughout this analysis, we are able to find that the Term Structure is still a valid predictor of economic activity, where Estrella and Hardouvelis believe this slope will be useful information not only for policymakers and investors, but in assessing the macro-economic environment (which extends beyond private investments and strategic measures) (Omrane et al). This is referenced in Graph 2, driven by the data (Data Output #2). The trend continues to show the predictive power in Graph 3, as introduced by the data in Data Output #3, and aligned with the literature introduced by Omrane et al.

The macro environment relies on these predictive trends, where term structure has proven that trends are going to continue, post-recession in this case. We could extend these findings even further, and analyze the findings in a recession, given all elements the same (Estrella and Hardouvelis). In doing so, we would predict that the trends are not diminished and hold their principle values, even though spending is going to be weakened and the value of the dollar is enhanced due to a lack of supply; these factors are consistent during a recessive period, and we appreciate this in coordination with Estrella and Hardouvelis’s introduction to predictive power under Term Structure (Omrane et al). The Appendix accounts for these factors in developing the data and supporting the visual representation (Graph 1, Graph 2, Graph 3).

Findings

As discussed, the findings show a similar picture where there is no data-driven evidence supporting PT over PO in terms of economic growth trends. PO began growth trends, where PT is going to sustain these trends. Estrella and Hardouvelis would look to this as a predictor of continued success or failure, rather than a dramatic change that has been brought on by PT with respect to the economy and the key variables. Here, the predictive scope of economics promoted by Estrella and Hardouvelis are proven, where PT has inherited an economic position (in a real world sense) and is sustaining this rather than changing the trends and adapting principles. PT has not done more for the economy than has been predicted, where the premise and hypothesis are going to hold true with this concern for term structure and the predictive variables, understood through Data Output #2 and Data Output #1 (Appendix).

Given the picture, it becomes clear that there is a break in the trend of GDP growth rates after the Recession. The official end of the Recession will correspond with the first instance. Here, we find that observed GDP growth rate is outside of the 95% CI of the growth rate prediction, based on the time period during the Recession. We can say with statistical significance that the GDP growth rate exceeded predicted growth rates, but this does not discount the structure of predictive measures; rather, this will align with the findings that the term structure is a real predictor of economic activity in a real world setting, as demonstrated with Data Output #3 (Omrane et al).

If the Recession would have continued, we would see expected unemployment growth to remain positive. Alternatively, the end of the Recession saw growth rates to be more negative than what would have been expected if the Recession were to continue. Here, the findings show that the growth rates would fall outside of the 95% confidence interval. Moreover, there are significant findings connected with unemployment growth rates. If PO were to continue beyond the term structure, we find that unemployment growth would be more negative than it was with PO causing a shift in January 2017. In forecasting the yield curve, predictive power accounts for the internal variables at the micro level to better-analyze the macro level (Omrane et al). Data follows this trends in accordance with the research findings, where Data Output #2 specifically holds reference to this trend.

Unemployment growth rate is not an indicator of term structure as being invalid. Particular segments of the population and demographics are going to cause a shift that goes against this predictive pattern, but does not take away from the real world application of this pattern. Certain segments of the population will either under or over-perform, and will go against this predictive trend. This is limited to a particular segment, and while these figures may fall outside of the 95% confidence interval, as is observed with unemployment growth rate, we still have confidence in the term structure and how it is an accurate predictor of real world macroeconomics.

Conclusion

The recession is going to shape the economy in an unfavorable fashion. Upholding predictive measures during a period of recession brings a new element, one that is less-supportive of fluidity and more-supportive of the value of tenacity in the economic scope. Real economics recognize these factors, and consider them in shaping the predictive forces moving forward. Here, the APPENDIX has accounted for these variables, where each understands that there is a recession and uses the data to account for any changes that would be brought on. Even so, the findings show consistency in the terms of the presidents (PO and PT), which are nothing more than symbolic of the economic trends once a country has faced a recession.

Continuing data that begins in past generations (1990s) and analyzing how it continues under the current framework shows predictive success. Estrella and Hardouvelis in “The Term Structure as a Predictor of Real Economic Activity” have demonstrated a successful measure, one that accounts for the scope of predictive measures in the macro environment, despite external factors that would have caused a shift. We have continued this trend with respect to a recession and a change in presidency. Despite claims that the economy was in a significant state of weakness under the guidance of PO, and is in a strong state based on PT’s initiatives, the data does not prove this. Instead, the data continues to show the practical and real world application of the term structure as a predictor of real economic activity, where the initial findings are confirmed, and the hypothesis is proven effective despite significant changes at the social and micro levels of economics. We can conclude with the term structure as being prominent and holding its value in the macro environment, with support of the data found in the Appendix as a guide moving forward. We expect these trends will continue to show support for the term structure and its inherent value in predicting the market. Term is a predictive variable, where we are able to predict future terms given a full assessment of the data, and with respect to Estrella and Hardouvelis achieving this practice through their proven measures and implications. Our Confidence Interval (95%) is appropriate as a Term predictor.

Data Output #1, Data Output #2, Data Output #3 demonstrate a trend under the predictive power of Term Structure. The Appendix presents a supplementary guide to the literature, showing the values that were implemented into the Graphs (Graph 1, Graph 2, Graph 3) under the predictive strategies of Estrella and Hardouvelis and their emphasis on current trends meeting the standards of recent trends, where our data has proven the trends to be accurate and valid in the macroeconomic environment. Sustainability is proven, where Term Structure holds true to its standards and supports the goals of the macroeconomic environment, within the chosen term (PO and PT).

Works Cited

  • Adjei, Frederick and Mavis Adjei. Economic Policy Uncertainty, Market Returns and Expected Return Predictability. Journal of Financial Economic Policy 9.3 (2017): 242-259.
  • DeJong, David N. and Dave Chetan. Structural Macroeconometrics. Princeton University Press, 2007.
  • Estrella, Arturo and Gikas Hardouvelis. The Term Structure as a Predictor of Real Economic Activity. Journal of Finance 46.2 (1991): 555-576.
  • Gurkaynak, Refet S. and Jonathan H. Wright. Macroeconomics and the Term Structure. Journal of Economic Literature 50.2 (2012): 331-367.
  • Omrane, Ben, Walid, He., Zhongzhi, Lawrence and Samir Trabelsi. Forecasting the Yield Curve of Government Bonds: A Dynamic Factor Approach. Managerial Finance 43.7 (2017): 774-793.
  • Orlik, T. Getting to Grips with China’s GDP Data. Upper Saddle River, NJ: Pearson Education, 2014.
  • Taylor, Mark P. Perspectives of Econometrics and Applied Economics. New York: Routledge, 2014.

Walton, G. M. History of the American Economy. Mason, OH: Cengage Learning, 2016.

APPENDIX

DATA OUTPUT #1

DATA OUTPUT #2

DATA OUTPUT #3

Cite This Work

To export a reference to this article please select a referencing stye below:

Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.

Related Services

View all

DMCA / Removal Request

If you are the original writer of this assignment and no longer wish to have your work published on the UKDiss.com website then please: