Magoosh GRE

Impact of Recession on Gold Prices

| March 18, 2015

The main aim of this study is to investigate the impact of recessions on gold prices. Several empirical findings have been published that assert that gold is usually a safe haven during recessions, and as a result, when the economy goes down, gold prices turn up. According to Saunders et al (2009), the methodology adopted in any research study depends highly on the research question and its associated objectives. This methodology chapter, would outline the research strategy, methods and data collection and analytical techniques that would be utilised in analysing the data necessary to answer the research question affirmatively, and without cause for invalidity or unreliability.

Remenyi et al (1998) postulates that the positivist philosophy entails working with an observable reality, and this usually leads to law like generalisations, as credible data could only be produced from data that can be observed or obtained. The main reason why the positive approach is being adopted – compared to the interpretivist approach, is that it requires the research strategy and questions to be based on already existing theories and developed hypothesis (Saunders et al, 2009), both of which constitute the major rationale behind this study. By adopting the positivist philosophy, data would be collected to test the research questions, so as to further develop the research background on the topic, investigate empirical findings, or suggest alternate hypothesis (Bryman and Bell, 2007). Saunders et al (2009) finally note that the positivist approach entails the adoption of an objective and external view, which the researcher perceives as crucial given the main aim of this study.

Two major approaches can be adopted while conducting research studies, and these are the deductive and inductive approaches. Deductive entails the use of existing research to formulate a research strategy and is usually effective in quantitative studies, whilst the inductive entails identifying a problem, then obtaining relevant answers through a purpose driven study (Oppenheim, 2005). Saunders et al (2009) note that the deductive approach is particularly useful in research strategies that are based on existing theories or empirical findings. Therefore the deductive approach would be utilised in this study in assessing the results from existing data, and analysing them against published findings.

In terms of the research strategy being adopted, a number of options are available for researchers, which include but are not limited to experiments, surveys, case studies, action research, grounded theory, and archival research. The three major strategies that are linked to the deductive approach are experiments, surveys and archival research. Experiments seek to find the relationship between 2 variables, with the researcher specifically determining the conditions of the variables, and controlling the conditions within which they occur. However, experiments are usually unfeasible for several businesses and management based research questions. Survey strategies are usually used to answer “who, what, where, how much and how many questions”, and they allow for collection of large amount of data from a sizable population in highly economical ways (Saunders et al, 2009 p. 144). However, the data collected by survey strategies are unlikely to be as wide ranging as those through other research strategies.

As a result of the disadvantages inherent in the experiment and survey strategies, and the fact that we cannot feasibly conduct surveys and experiments on gold prices and GDP growth rate over the past 30 years, this study would be adopting an archival based research strategy. Babble (2008) describes archival research as that which incorporates the use of administrative records and documents as the principle source of data. They often incorporate time series data that are utilised in answering economic related questions (Oppenheim, 2005). Therefore since this study aims to investigate how recession has impacted gold prices, the most ideal research strategy would be the archival research.

The utilisation of archival research allows this study to answer questions that are directed towards the past and changes that have occurred over time, and even allow several time series variables to be collected. The researcher’s ability to answer the research question is however dependent on the nature of the data collected. Common disadvantages of this approach are that archival research tends to be descriptive or explanatory, and some data may be missing. Furthermore, the data has not been collected for the sole purpose of answering the research question – compared to survey and experimental methods. This could impact on the quality of the analysis, and the manner in which the variables are analyzed to answer the research question. This study would seek to circumvent the disadvantages inherent in the archival strategy by only seeking data that directly relates to the research question – gold prices and recession. Archival strategy was also chosen as it was the same strategy adopted by Starr and Tran (2008) and Shaflee and Topal (2010), in which the deductive approach utilised in this study is based on.

Furthermore, previous studies would be consulted, in order to find ideal variables that have been used in comparing recession against gold prices. For instance, recession is a generic term that may mean a number of things. Therefore, previous studies on both phenomena, such as those conducted by Starr and Train (2008) and Hamlin (2006), have used specific variables to calculate recession. These would be outlined in the section on data collection and analysis.

Quantitative and Qualitative methods are two methods mainly used in the data collection process of research. The former involves data that is either in the form of, or expressed as numbers while the latter involves collecting data in the form of words (Easterby-Smith et al, 2008). According to Saunders et al (2009), quantitative is usually used as a synonym for data collection techniques such as questionnaire or data analysis engagements that use or generate numerical data. Qualitative data however is utilised as a synonym for data (such as data categorisation) that leads to the use or generation of data that is non-numeric. Qualitative data is however not limited to words, it could also refer to videos and pictures (Saunders et al, 2009).

Tashakkori and Teddlie (2008) further note that both methods need not used in isolation, and researchers conducting a study could choose to use one of either methods (mono method approach) or a combination of both methods (multi method approach). In the mono method approach, the research is based on either a qualitative or quantitative data studies. The major merits of quantitative data is that is allows for easier analysis, and results found are usually generalisable, reliable and valid, given that the analytical approach are stated.
Quantitative data are also attributed as being more objective, not easily open to researcher or respondent bias, and can be readily analyzed and compared to previous or future data (Oppenheim, 2005). Furthermore, quantitative data, when analyzed appropriately can successfully ascertain the relationship and / or correlation between one or more variables, and also depict the extent to which one variable could affect the other (Saunders et al, 2009). However, Saunders et al also note that quantitative studies do not fully explain all the factors affecting a variable. The figures alone do not account for human factors, and cannot accurately predict how changes within the “environment” would affect how that value would fare in the future.

Qualitative studies on the other hand are more subjective in nature, and deal with research studies that are more attuned to the human aspect of situations. For instance, management research on factors such as motivation, performance management, and organisational change, usually tend to be qualitative in nature (Saunders et al, 2009). This is due to the fact that the researcher would need to take into account the respondents’ view, and also because the researcher may be involved in the study in one of several ways. Data obtained through qualitative studies are usually in words, and help to provide insight into factors underlying particular situations. Bryman and Bell (2007) note that qualitative studies are usually utilised in inteprevitist philosophies, as it aids the researcher in interpreting social phenomena.

Multi method refers to those combinations where more than one data collection technique is used with associated analysis techniques (Tashakkori and Teddlie, 2008).

This choice of adopting a multi method approach is increasingly advocated within business and management research (Curran and Blackburn, 2001), where a single research study may use quantitative and qualitative techniques in combination as well as use primary and secondary data (Saunders et al, 2009). However multi methods are usually restricted within either a quantitative or qualitative view (Tashakkori and Teddlie, 2008). Therefore the researcher’s decision of choosing to collect quantitative and qualitative data using questionnaires whilst analysing these data using statistical procedures is a multi-method quantitative study (Saunders et al, 2007). Tashakkori and Teddlie assert that multiple methods are useful because they provide better opportunities for the researcher to answer questions and they also allow more efficient evaluations to which research findings could be trusted.

This study would be adopting a quantitative view to data collection, as Saunders et al (2009) recommends that deductive studies adopting an archival strategy are most likely quantitative due to the ease of finding and analysing particular variables that would be crucial in answering the research question. The data sought for this study are data on gold prices and recession, all of which are quantitative, and would allow for effective analysis using explanatory analysis. Quantitative analysis also allows for effective comparison between one or more variables, thus illustrating its significance when comparing the prices of gold to recession figures in the country being analyzed.

This project is being modelled based on empirical research conducted by Starr and Tran (2008), in which they compared gold prices over the past 22 years to recession figures. According to Lester (2009), recession represents negative growth rate. The National Bureau of Economic Research (2001) however define as a significant decline in economic activity lasting more than 2 months, visible in employment, real income, wholesale and retail trade, and industrial production. As a result this study has identified gold prices as the dependent variable, while recession figures such as unemployment rate, inflation rate and GDP growth rate as the independent variables, as they directly relate to Lester’s definition of a recession.

Time series quantitative data would need to be obtained on all four variables in order to effectively ascertain how recessions impact gold prices. For simplicity, only one country would be analyse: UK, and the timeframe depicted is 40 years, to account for historical rises and falls in core economic and gold price variables over the past 4 decades. Annual averages for gold prices in GBP from 1970 – 2010 have been obtained from the World Gold Council (, while data on UK GDP growth rate, unemployment figures, and inflation figures have been obtained from the Office of National Statistics ( Specific data on just the variables required, that corresponded to the years being analyzed (1970 – 2010) were copied from each dataset in to a new dataset in Microsoft Excel.

The quantitative data collected during the course of this study, whilst still in its raw form, is described by Saunders et al (2009) as being useless and conveying little information to most people. The data on all four variables has therefore been analyzed using descriptive and analytical tools. The first major analysis conducted was a descriptive chart that was drawn on all four variables over the time period, to graphically illustrate how well all four variables have fared over the past 40 years. Correlation studies were then conducted in Excel in a bid to determine how well the movements in one variable correlated with other variables. The correlation coefficient showed how significant movements in any 2 variables were, and are depicted in the following chapter.

Finally, the variables that did tend to correlate with movements in gold prices, where then utilised in conducting a multiple regression analysis. Regression analysis is a process focusing on discovering relationships between a dependent variable and any number of independent variables. With regression analysis the researcher can understand how changes to an independent variable affect a dependent variable, thus the relationship between the two (Oppenheim, 2005). Only those variables that showed a positive or negative correlation value significant enough beyond the 5% limit were utilised in the regression analysis, based on recommendations by Oppenheim (2005), who argued that regression analysis are only useful when participating variables have a high positive or negative linear relationship. The regression analysis was used to determine the extent to which these variables determined changes in the gold price, a value otherwise known as the conditional expectation for the price of gold.

The main aim of this research project has been to ascertain how recessions affect the price of gold. Several empirical studies have been conducted on the same phenomenon and most have found that adverse effects on economies of particular countries, especially western countries, has a beneficial effect on the price of gold (Starr and Tran, 2008; Wang et al, 2010; and Hamlin (2006). In periods of economic recession, these studies have found that gold prices tend to fall. It has been broadly regarded as a safe haven for investors the world over. However, the sales of gold are not solely based on economic recession, as several factors such as the growth in industry also positively affect gold prices. World Gold Council (2011) noted that several industries, including the healthcare industry, manufacturing industry, and even dentist utilise gold in their operations, it is therefore invalid for the price of gold to be based solely or largely on economic recessions.

This study has therefore opted to investigate these theories, particularly the findings of Baur and McDermott (2010) and Starr and Tran (2008). Both studies were conducted using fiscal data on several countries, against the price of gold over a time series period of 30 years. The findings from this study illustrated that gold may be both a hedge and safe haven for major European stock markets, and the US, not but in emerging markets. On the contrary, the demand for gold in emerging markets has been on the rise lately due mostly to their economic prosperity (Stanczyk, 2010). In a bid to investigate such theories, this study has focused on the UK, given limitations in gathering and analysing broad time series data on several countries. 40 year Gold prices (in GDP) from 1970 – 2010, has been utilised as the main dependent variable, while inflation, unemployment rate and GDP growth figures – all of which have been ascertained by Lester (2009) to be indicators of recession – have been utilised as the main independent variables.

Results from the descriptive analysis – in which the figures of all variables were plotted on a line graph – show that while inflation has generally subsided over the past 30 years, gold prices have been on a steady rise growing by over 5000% from £15.03 in 1970 to £792.51 in 2010. GDP growth rate has generally had a range of -2% – 2% over the past 40 years as well. Unemployment rate has also generally been between 5 – 10%, with recent figures showing 7.7% in 2009.

The major trends immediately noticeable from the charts are that gold prices have risen considerably, while inflation has dropped. Correlation studies were conducted on all four variables, in a bid to see how well movements in one variable did correlate with similar or alternate movements in the other variables, and the results are depicted in figure 2. The results are sorted based on the degree of correlation, and it shows that gold prices and inflation rates are the most correlated variables within the UK. Their relationship is such of a negative correlation, illustrating that as inflation rates go down, gold prices go up – which is also noticeable from Figure 2. The findings depict a significance level of below 0.01.

@22@21only be wrong 1% of the time. While that between gold prices, unemployment and GDP growth illustrate that the figures could be wrong only 5% of the time. The implications of these findings, based on the variables collected on the UK, show that a rise in gold prices inversely correlates with a rise inflation, and GDP growth, and positively correlates with rise in unemployment rates. A linear scatter diagram between gold prices and inflation were plotted, as seen in figure 3, and the results show a highly negative correlation, with a coefficient of determination of 18% (showing that 18% of the changes in gold prices may be explained through change in inflation rates). When recent data from 2007 – date were taken out, the coefficient of determination was over 24%. @!21As a result of all variables being highly correlated, multiple linear regression analysis were conducted with gold prices as the dependent variable, as shown in figure 4, and all other variables as the independent variables. The results show a high correlation between movements on all variables, showing a correlation coefficient of 0.81867. The coefficient of determination, which aims to estimate the ratio to which these values could impact on movements in gold price is 0.67, thus illustrating that 67% of the movements in gold prices could be explained through movements in inflation rates, unemployment and GDP growth rate.@11

The analyses of these variables have been conducted in line with methodologies used in several empirical studies, and the data on gold prices movement do show an astronomical increase. However, since the theory is that gold prices move inversely to economic situations in western countries, then an explanation could be that the steady increase in gold prices are as a result of steady economic decline across all major economies over the past 40 years. However, that is not the case, which shows that even though gold prices are somewhat related to economic recessions, they do not represent the total rationale in gold price movements (World Gold Council, 2011).

However, the findings from this study do give credence and validity to the previous empirical findings. The correlation studies, which show that increase in gold prices has been linked with an increase in unemployment and decrease in GDP growth, validate findings of Baur and McDermott (2010) who found that gold is usually considered a safe haven in major European markets. The most significant results lie in the inverse correlation with inflation, which goes in line with explanations by Caple et al (2005). Hamlin (2006) noted that gold prices usually go opposite of inflation rates. As inflation rates increase, gold prices fall, because gold does not bear interest. Gold prices also tend to increase when inflation rates fall. Inflation rates are highly correlated with interest rates (Lester, 2009).

Therefore, based on these findings, it is apparent that in the UK, gold prices are negatively correlated to inflation figures as depicted by Lester (2009); goes up as unemployment rates go up (Baur and McDermott, 2010), and goes up as GDP growth rates decline (Starr and Tran, 2008). The multiple regression analysis also provides credibility to the findings, as it shows that all three independent variables could account for 67% of the movements in gold prices.

However, correlation does not necessarily imply cause, and even though economic effects have an impact on gold prices, gold is also used for a number of other factors, which grow in value as the economy grows – such as manufacturing processes (World Gold Council, 2011). Furthermore, this study was only conducted on the UK but the prices of gold are international, therefore it is assumed that several other economies have an equal impact on the price of gold and not just the UK. In order to effectively ascertain the international effect of economic recession on gold prices, an international time series study would need to be conducted, such as that of Starr and Tran (2008), on several countries and the individual and collective impact of their economies on the price of gold.

This study has investigated the relationship between recessions and gold prices, with time series data from the UK. Data on gold prices, GDP growth rate, inflation rates and unemployment rate were gathered through relevant sources, and analyzed. Correlation analysis showed that gold prices were inversely correlated with inflation rates, and GDP growth rates, but positively correlated with a rise in unemployment. A fall in GDP growth rate, fall in inflation and a rise in unemployment are considered tell-tale signs of a recession, so based on this assumption, this study confirms that gold prices go up as certain economies experience an economic downturn.
Arnold, R. (2008) Economics. 9th edition. South-Western Cengage Learning: Mason, OH.

Baur, D., and McDermott, T. (2010) ‘Is gold a safe haven? International evidence.’ Journal of Banking and Finance 17,2 91-99.

Bryman, A. and Bell, E. (2007) Business Research Methods, 2nd Ed, Oxford University Press: Oxford, 786pp

Cai, J Y-L C and Wong, M (2001) ‘What Moves the Gold Market?’, Journal of Futures Markets, 21:3, 257–78.

Capie, F, Mills, T C and Wood, G (2005) ‘Gold as a Hedge against the Dollar’, Journal of International Financial Markets, Institutions and Money, 15:4, 343–52.

Hamlin, J. (2006) ‘Impact of Recession on Gold.’ Investment Strategies, September 2006.

Knoop, Todd (2010) Recessions and depressions: understanding business cycles. Greenwood Publishing Group: Santa Barbara: CA.

Lester, A. (2009) Growth Management: Two Hats Are Better Than One. Palgrave Macmillan: Hampshire.

Office for National Statistics (2011) UK GDP, Inflation Rate and Unemployment figures,, [accessed: 27/04/11]

Oppenheim, A. N. (2000) Questionnaire design, interviewing and attitude measurement, Continuum: London, 303 p

Phillips, J. (2006) ‘Where Does Gold go in a U.S. Recession.’ Financial Sense, September 2006.

Robson, C. (2002) Real world research: a resource for social scientists and practitioner- researchers, 2nd Ed, Wiley-Blackwell: NY, 599pp

Saunders, M., Lewis, P. and Thornhill, A. (2007) Research methods for business students, Pearson Education: NJ, 624pp

Shaflee, S., and Topal, E. (2010) ‘An overview of global gold market and gold price forecasting.’ Resources Policy 35,3 178-189.

Shiskin, J (1975) ‘The Changing Business Cycle.’ New York Times.

Stanczyk, A. (2010) ‘Precious metals come in as top asset four out of five years, and the top asset of 2010 according to Lloyds TSB Assetwatch.’ Telegraph, March 2010.

Starr, M and Tran, K (2008), ‘Determinants of the Physical Demand for Gold: Evidence from Panel Data’, The World Economy, 416-436.

Tashakkori, A. and Teddlie, C. (2008) Foundations of Mixed Methods Research: Integrating Quantitative and Qualitative Approaches in the Social and Behavioral Sciences, Sage Publications: UK, 400pp

Wang, K., Lee, Y., and Thi, T. (2010) ‘Time and place where gold acts as an inflation hedge: An application of long-run and short-run threshold model.’ Economic Modelling, December 2010.

World Gold Council (2011) Investment – Gold Prices, , [accessed: 28/04/11]
5 APPENDIX@12_001@13@14@15

Category: Business Essay Examples, Essay & Dissertation Samples, Finance Essay Examples