Abstract
Background: Analysing the persistence of consumption patterns is fundamental to identifying latent economic dynamics and the underlying factors of market fluctuations. In the BRICS Plus countries, this involves a complex interplay of socioeconomic, cultural and political conditions.
Aim: To study the degree of consumption persistence in 11 countries to determine whether shocks have transitory or permanent effects. Additionally, we examined how external shocks may have impacted spending patterns there.
Setting: The data used are ‘Final Consumption Expenditure of Households and NPISH’ from the World Bank and are annual (1960–2022).
Method: Fractional integration techniques. These econometric methods analyse time series with long-term memory, meaning that past events influence the future over an extended time horizon, although they fade over time.
Results: There are substantial differences. The most significant positive trends are in Brazil, Iran and Russia, as well as Ethiopia and Saudi Arabia. Only Iran and Saudi Arabia show evidence of mean reversion and temporary shocks. For the other countries, we cannot rule out the possibility that the shocks are permanent, as the unit root hypothesis is not rejected.
Conclusion: Each country exhibits distinct strengths and weaknesses. The methodology helps us analyse past consumption and its persistence to understand future consumption. Policy recommendations include fiscal stimulus, increased disposable income, investment in infrastructure, among others.
Contribution: No articles or books were found that analyse this topic with fragmented integration in Brazil, Russia, India and China (BRICS) Plus countries. Furthermore, the results are heterogeneous between countries, implying that the same policy recommendations cannot be generalised.
Keywords: consumption persistence; BRICS Plus; shocks; fractional integration; policy recommendations.
Introduction
The comprehension of consumption behaviour dynamics holds paramount significance for various stakeholders, including researchers, economists and policymakers committed to unravelling the intricacies underlying long-term economic growth and macroeconomic fluctuations. As a crucial element of aggregate demand, consumption reflects household purchasing power while simultaneously playing a primary role in determining the macroeconomic performance of a country or territory. Consequently, analysing the stability and continuity of consumer spending habits is vital for identifying hidden economic forces and understanding the underlying causes of economic fluctuations (Kearney 2010; Warde 2014; etc.) This study concentrates on an inquiry into the durability of consumption behaviour throughout the diverse landscapes of the different countries that constitute the BRICS Plus, marked by varying socioeconomic conditions, cultural influences and policy frameworks.
The term ‘BRIC’, originally excluding South Africa, was coined by O’neill (2001), then the chief economist at Goldman Sachs. In 2009, these four countries (China, India, Russia and Brazil), proposing to reach and equal the North American and Western countries in the global order, came together to form a semi-institutionalised political outfit. We should note that BRICS is not a formal international organisation like the World Bank, Organization of the Petroleum Exporting Countries (OPEC), or the United Nations. The leaders of these countries, including heads of state and government, gather annually, with each nation taking a 1-year turn as the rotating chair of the group. They discuss matters of mutual interest, and their collective influence has grown in areas such as trade, investment and global governance. South Africa, despite being the smallest member in terms of economic strength and population, became the first addition to the BRIC group in 2010, leading to the new acronym, BRICS. Together, these five countries contribute 25% of the global economy’s output and represent over 40% of the population in the world (Stuenkel 2020). In August 2023, during the Johannesburg convention, it was decided that another six countries could join the organisation. These were: United Arab Emirates (UAE), Saudi Arabia, Iran, Ethiopia, Egypt and Argentina (Reuters 2023b). The new name for the 11 countries is BRICS Plus as China proposed and their membership began in 2024 (The Business Standard 2023).
The main objective of this work is to analyse the degree of persistence in consumption in these 11 countries to understand if shocks have transitory or permanent effects. For this analysis, we use fractional integration techniques, which are quite flexible and more general than the standard methods based on unit roots or stationarity tests.
To present an economic study with theoretical foundations, we cannot overlook the great thinkers who contributed their knowledge to our current understanding of persistence in consumption. Their contributions are summarised in Table 1.
| TABLE 1: Comparative summary of seven theories of consumption. |
The theories mentioned in Table 1 will be briefly described in the literature review section.
The subsequent parts of this article are organised in this way: after outlining the Contextual setting to the BRICS Plus countries, we provide a comprehensive Literature review of the pertinent and most recent articles concerning consumption behaviour in general, consumption in the BRICS nations, external shocks in the BRICS countries and fractional integration. We then elucidate the chosen Methodology, offering a detailed exposition of the fractional integration framework, before delineating the dataset, encompassing the 11 different countries that form the basis of our analysis. Through the Results section we present the empirical findings, delving into the persistence characteristics of consumption while identifying potential variations specific to each country. In the Conclusion we recapitulate the key findings and proffer recommendations with regard to policy implications across the BRICS Plus nations.
Contextual setting
The BRICS countries formed the alliance to enhance cooperation on economic, political and strategic issues.
While each BRICS country has its unique challenges and opportunities, they as a collective configure an important section of the world’s population, geopolitical influence and economic output, as we can see in Figure 1.
 |
FIGURE 1: Map of the original BRICS members and the additional members. |
|
Relevant data for these countries are summarised in Table 2, showing that these countries are still developing, but they can bring together synergies that help their growth.
| TABLE 2: Data for the BRICS Plus countries in 2022. |
The data indicate that Saudi Arabia boasts the most favourable conditions among the countries analysed, ranking first in NMW salary, GDP and standard of living. It holds the second position in CPI, following China, and secures the fourth spot in the economy ranking, trailing behind China, India and Russia – all significantly larger nations. On the contrary, Ethiopia, as per the data, exhibits concerning indicators of underdevelopment.
Examining the member countries of BRICS Plus, it is noteworthy that many of them are major oil producers, with six of them ranking among the top 10, as it is showed in Table 3.
| TABLE 3: The top 10 oil† producers and share of total world oil production‡ in 2022§. |
These data led to an incorrect conclusion when new members joined, as it was initially believed that they could collectively reach 80% of the global petroleum production. However, the actual scenario indicates that this figure will remain below 50%, which, nevertheless, remains a noteworthy statistic (Reuters 2023c). This circumstance may entail economic influence, influence on market dynamics, reduced dependency on other countries (in terms of energy), and also have environmental implications.
In the future, the possibility exists for additional countries to join the organisation. As reported by the 2023 summit chair, South Africa, over 40 countries, including Algeria, Bolivia, Indonesia, Cuba, the Democratic Republic of Congo, Comoros, Gabon and Kazakhstan, have conveyed their interest in participating in the organisation (Reuters 2023a).
By employing fractional integration techniques, our objective is to elucidate the enduring memory characteristics of consumption conduct, thus providing invaluable insights into the evolution of household expenditures across the years.
In order to ensure thorough coverage and enable cross-country comparisons, our study examines an extensive dataset spanning 11 countries. This wide-ranging approach allows us to include a diverse variety of cultural characteristics, economic configurations and regulatory environments, which provides a solid empirical base for studying how persistently consumption behaves across these jurisdictions. Consequently, we are able to analyse potential differences in consumer habits that stem from country-specific factors, including societal values, income tiers and institutional structures.
The outcomes of our research bear substantial implications for policymakers, offering valuable insights into the resilience and adaptability of consumption trends throughout the BRICS Plus territory. Moreover, it sheds light on the efficacy of policy measures aimed at invigorating economic growth, mitigating income disparities and promoting sustainable development (De Haan, Lundström & Sturm 2006; Rodrik 2005). Our overarching objective is to gain a comprehensive understanding of prospective shifts in these countries’ consumption patterns and ascertain whether enduring patterns exist within the data series under examination.
Literature review
The examination of consumption is far from a recent endeavour, and there exists a substantial body of research on this subject. Regarding the specific topics, in 2023 alone, we find papers dealing with the following issues: energy consumption and carbon emissions forecasting (Hu & Man 2023); energy consumption and potential renewable sources (Hassan et al. 2023); meat consumption and risk of ischaemic heart disease (Papier et al. 2023); the transaction behaviour of cryptocurrency and electricity consumption (Zheng et al. 2023); the influence of economic factors on sustainable energy consumption (Sadiq et al. 2023); future warming from global food consumption (Ivanovich et al. 2023); climate change impacts on future residential electricity consumption and energy burden (Jones et al. 2023); energy consumption and its influence on the policy drive towards Sustainable Development Goal 7 (Elavarasan et al. 2023); inflation with Covid Consumption Baskets (Cavallo 2024), to mention only a few.
Focusing on theories that justify persistence in consumption, there are different studies. Keynes (1937) provided an early foundation for consumption persistence through his ‘fundamental psychological law’, noting that a household’s habitual standard of living has the primary claim on its income. This behavioural rigidity implies that consumers do not alter their spending immediately following income fluctuations, but only after a notable lapse of time. By establishing that consumption is inherently sticky and lags behind sudden economic shifts, he provided the core theoretical rationale for short-term momentum, explaining why aggregate spending patterns persist over time rather than adjusting instantaneously to shocks. A few years later, Boulding (1945) reminded us that consumption is fundamentally the depreciation of an accumulated stock of goods. Because durable goods physically persist across multiple time periods, the utility they provide is inherently distributed over time, establishing a foundational structural reason for why consumer behaviour cannot be analysed purely as a series of isolated, short-term transactions. Duesenberry (1948) formalised consumption persistence through the ‘ratchet effect’ within his relative income hypothesis. He argued that current expenditure is heavily driven by a household’s historical peak income rather than just its current earnings. Because consumers adapt to and fiercely protect their highest attained standard of living, they resist scaling back their spending during economic downturns, choosing instead to absorb negative income shocks by reducing their savings. This asymmetric behaviour creates a powerful structural momentum, demonstrating how past peak consumption acts as a historical anchor that prevents immediate mean reversion and locks in high expenditure patterns over time. Brown (1952) introduced the concept of ‘habit persistence’ into aggregate consumption functions. Moving away from static framework models, Brown theorised that past consumption experiences cast a continuous psychological shadow over current expenditure decisions. By structurally incorporating lagged consumption as an explanatory variable, his framework demonstrated that customs, traditions and acquired standards of living create a ‘hysteresis’ effect. This seminal contribution provided the foundational econometric justification for time lags in consumer behaviour, establishing that current consumption is inherently a path-dependent process heavily dictated by historical consumption trajectories. Friedman (1957) rationalised consumption persistence through his permanent income hypothesis, which posits that households base their spending on long-term expected wealth rather than current income. Because consumers use savings and credit to smooth out transitory income shocks, aggregate consumption remains highly stable and decoupled from temporary fluctuations. This forward-looking behaviour imparts a structural momentum to the data, as current consumption is anchored to a gradually updated measure of permanent income, explaining why spending patterns exhibit strong continuity and resist immediate changes following economic disturbances. Hall (1978) provided a foundational, albeit contrasting, framework for consumption persistence by introducing the rational expectations ‘random walk’ hypothesis. He demonstrated that if consumers optimally smooth their spending based on all available information, current consumption incorporates all past knowledge, making its future changes completely unpredictable. Under this strict theoretical framework, consumption does not exhibit a slow, lagged adjustment to past events; rather, any shock to permanent income shifts the consumption path immediately and permanently. Hall’s model thus formalises persistence in its most extreme statistical form – a pure unit root (I(1)) process – where the level of consumption inherits the entire history of past structural shocks, which never fade or revert to a mean.
Browning and Crossley (2001) justified consumption persistence by evaluating the modern, microfounded life cycle model, emphasising that apparent behavioural inertia often reflects rational, forward-looking smoothing under market frictions. They highlighted that when households face adjustment costs, habit formation or liquidity constraints, the marginal utility of consumption becomes linked across time periods.
Rather than adjusting expenditures instantly to every economic disturbance, consumers deliberately smooth their spending path to avoid costly disruptions to their lifestyle. Consequently, their framework demonstrates that persistence is not merely an irrational delay, but a structural feature of optimal intertemporal allocation, explaining why aggregate consumption patterns exhibit long-term continuity and a delayed response to shocks.
Based on the above theories and focusing on the methodology, the fractional integration techniques used in this work capture long-memory processes and hyperbolic shock decay. This econometric behaviour finds its closest theoretical counterpart in Brown’s (1952) Habit Persistence framework, which posits that acquired living standards create an enduring behavioural inertia, causing aggregate consumption to adjust to economic shocks at a gradual, protracted pace.
Shifting our attention to articles addressing household consumption in BRICS, we have not identified any publications on this subject. Despite this, our research reveals that over the past 5 years, there has been a notable trend in the discussion of energy consumption (Ibrahim et al. 2022; Muhammad et al. 2021; Nawaz et al. 2021; Sachan et al. 2023; Samour et al. 2023; Wang, Bui & Zhang 2020; etc.).
There are studies of the impact of external shocks on different topics in the BRICS nations. An et al. (2022) assessed the probability of an external liquidity crisis by evaluating the early warning system (EWS) metric via an ARMA-Generalized Autoregressive Conditional Heteroskedasticity (GARCH) modelling approach. The results reported that South Africa, India and Brazil are highly vulnerable to shocks, whereas Russia and China present a more reasonable level of risk in the near future. Batondo and Uwilingiye (2022) examined the co-movement of stock markets in BRICS countries and the United States (US) using daily close price indices of selected stock markets from 2000 to 2021. They applied a wavelet decomposition on stock return series. The results showed that most of the majority of preceding shocks were transmitted through mutual reliance resulting from trade channels and advancements in economic integration.
In conclusion and from the perspectives of risk management and portfolio diversification, the Chinese market offered highly profitable opportunities for short-term investors from the remaining nations examined in this research. Wen et al. (2022) examined the asymmetric and heterogeneous effects of monetary policy uncertainty (MPU) on stock returns in the BRICS countries and in the Group of Seven (G7) countries using a quantile-on-quantile (QQ) approach. The conclusion reveals that the market reactions to MPU shocks across the G7 nations appear more erratic than those observed within the BRICS economies. Al-Mohamad, Jreisat and Sraieb (2023), using Diebold and Yilmaz’s (2012) approach, measured the spillover transmission and the financial connectedness among stock markets between BRICS economies and the rest of the world in the wake of the Ukrainian crisis with Russia. The outcomes of the volatility transmission demonstrated that stock markets in BRICS became less resilient to external shocks (volatility shock) as compared to the past. Su et al. (2021) explored the effect of crude oil price (COP) shocks on the economic policy uncertainty (EPU) in BRICS countries using the quantile Granger causality test. The empirical findings of the causal relationship exhibited asymmetrical characteristics, meaning that when supply shocks occur in the oil markets, they have a beneficial impact on the EPU in Brazil, China and India. Furthermore, a decreasing COP affects the EPU in South Africa and Russia.
Regarding our chosen methodology, fractional integration, it should be noted that several different studies can be found that have employed this approach to investigate a wide range of subjects. These include, but are not limited to: analysing US aggregate output dynamics (Diebold & Rudebusch 1989); examining interest rates on long-term government bonds issued by central governments in the US, Canada, Germany, the United Kingdom and Japan (Barkoulas, Baum & Oguz 1996); investigating inflation patterns in 10 different countries (Baillie, Chung & Tieslau 1996); studying the stochastic behaviour of unemployment in 11 African nations (Caporale & Gil-Alana 2018); assessing the efficiency and persistence of volatility in 12 cryptocurrencies (Yaya et al. 2021); understanding the impact of coronavirus disease 2019 (COVID-19) on Turkey’s tourism sector (Yucel et al. 2022); analysing US car sales (Lopez & Gil-Alana 2023); among others. These examples illustrate the versatility and effectiveness of fractional integration as a research method across diverse fields and topics.
In all the literature presented above, we have not found any single article or book that analyses the topic examined in this work with fractional integration, which makes this investigation in the BRICS Plus countries potentially of interest in the field.
Methods
In our context, we designate L as the lag operator, signifying that Lkx(t) equals x(t-k). A series is considered to be integrated of order 1, denoted as I(1), if it can be formulated in this manner (Equation 1):
In this context, x(t) denotes a time series, and u(t) is a covariance stationary process with short memory, also denominated integrated of order 0. Short memory processes exhibit a distinctive trait – the infinite sums of their autocovariances, expressed as γ(u) = E[(x(t) – μ)(x(t + u) – μ)], are finite, signifying that (Equation 2):
Within the category of short memory processes, we can encompass the traditional AutoRegressive Moving Average (ARMA) processes. Specifically, if u(t) follows an ARMA (p, q) pattern, then x(t) is characterised as ARIMA (p, 1, q). As detailed in the preceding section, numerous statistical testing methods exist to ascertain whether a series is I(1) or I(0), also known as short memory. These methods include ADF (Dickey & Fuller 1979), PP (Phillips & Perron 1988), KPSS (Kwiatkowski et al. 1992), ERS (Elliot, Rothenberg & Stock 1996) and NP (Ng & Perron 2001), yet they all exclusively consider integer orders of integration (i.e. 1 for non-stationary series and 0 for stationary ones).
Conversely, scholars like Diebold and Rudebusch (1991), Hassler and Wolters (1994), Lee and Schmidt (1996), among others, have demonstrated that many of the aforementioned approaches exhibit limited effectiveness when the true data generating process adheres to I(d) where d represents a fractional value. If the differencing parameter d is fractional, Equation (1) is expressed in the following way (Equation 3):
Moreover, when d takes on a positive value in Equation (3), x(t) in (3) transforms into a long-memory process. This implies that the infinite sum of autocorrelations becomes unbounded, that is (Equation 4):
In alternation, we can characterise the two categories of processes: short memory and long memory, in the frequency domain. By formulating the spectral density function, f(λ), as the Fourier transform of the autocovariances, that is (Equation 5):
we categorise x(t) as having long memory when the spectrum extends to infinity at least at one point within the frequency range [0, π) (Equation 6):
Conversely, it is classified as short memory if the spectral density function, f(λ), is positive and limited across all frequencies (Equation 7):
Similar to the scenario of white noise, where f(λ) remains constant across all values of λ, it can be demonstrated in the context of Equation (3) that the spectral density function of x(t) is (Equation 8):
and it tends to infinity as λ → 0+ with d > 0 justifying its long memory property.
The fractional integrated process outlined in Equation (3) provides a diverse range of specifications, encompassing, among other possibilities:
- anti-persistence, if d < 0
- short memory, if d = 0
- long memory, though covariance stationary processes, if 0 < d < 0.5
- non-stationary mean reverting processes, if 0.5 ≤ d < 1
- unit roots if d = 1
- and explosive processes if d > 1.
Furthermore, the model in Equation (3) can be expanded to accommodate linear, or even non-linear, trends. In a conventional approach, aligning with established unit roots literature (Bhargava 1986; Schmidt & Phillips 1992), we incorporate a linear time trend model structured as follows (Equation 9):
Here, α and β denote the parameters requiring statistical estimation – specifically, an intercept and a linear time trend, respectively. The variable x(t) is determined by equation (3).
In the practical application discussed in the subsequent section, our preliminary analysis relies on a framework delineated by Equations (9) and (3). We start from the premise that the error term u(t) follows a white noise process, with zero mean and constant variance.
The estimation of the parameters is conducted via maximum likelihood in the frequency domain and using a simple version of a testing procedure developed in Robinson (1994). It tests the null hypothesis: Ho: d = do, in the model given by the Equations (9) and (3), where do can be any real value. Using this procedure, which has a standard N(0, 1) limit distribution, we are able to compute a confidence band of values where do cannot be rejected (see, e.g. Gil-Alana & Robinson 1997 for the description of the functional form of the test statistic used in this application).
Data
The data utilised in this analysis are sourced from The World Bank (2023b) and pertains to ‘Households and NPISHs Final consumption expenditure’ measured in current US dollars. NPISHs stands for Non-Profit Institutions Serving Households. In simpler terms, the data represent the total spending by NPISHs and households on various goods and services that they use for themselves, such as housing, food, healthcare, transportation, leisure activities and education. It is worth noting that the data are not corrected for inflationary effects or temporal shifts in purchasing power, and it offers insights into the spending patterns and trends of households across BRICS Plus countries from 1960 to 2022.
Additionally, NPISHs constitute institutional units that are separate from public sector bodies and commercial enterprises. These entities are engaged in non-profit actions and supply various services and goods to households and also communities. Non-Profit Institutions Serving Households are charities, NGOs, religious institutions, foundations, volunteer organisations and community centres. They perform a critical function in meeting societal demands, advancing public well-being and fostering the development of communities.
In the realm of economics and demographics, households denote collectives of individuals who live all together in a shared home and engage in economic and social activities collectively. Understanding households, including aspects like their income levels, spending habits, demographics and socioeconomic conditions, is crucial for researchers, policymakers and businesses across different fields, ranging from sociology and economics to public policy and marketing.
Figure 2 illustrates a significant positive trend in the data for the World in general, indicating substantial development over time. This development is evident in the increase in total expenditure from US$1.75 trillion in 1970 to US$53.1 trillion in 2021. However, not all countries follow the same trend, and it is important to analyse and compare these results to understand the differences between them and the degree of persistence in their expenditure patterns. The data include information about 11 countries: Argentina, Brazil, China, Egypt, Ethiopia, India, Iran, Russia, Saudi Arabia, South Africa and United Arab Emirates. Certain nations lack data extending back to 1960, and this limitation is explicitly documented in Table 4.
 |
FIGURE 2: Households and Non-Profit Institutions Serving Households final global consumption expenditure (current US$). |
|
Ethical considerations
This article followed all ethical standards for research without direct contact with human or animal subjects.
Results
The model under consideration is the one outlined by Equations (9) and (3), specifically (Equation 10):
The results are presented in Table 5 and Table 6. Table 5 presents the estimations of the parameter d in three traditional scenarios from the unit root literature:
- Without deterministic terms, assuming both unknown coefficients, α and β, are pre-set to zero in (10).
- With an intercept (β = 0).
- With both an intercept and a linear time trend.
| TABLE 5: Estimates of the fractional differencing parameter. |
| TABLE 6: Estimated coefficients of the selected models. |
The values highlighted in bold in Table 5 correspond to the chosen specifications for each regressor type, and the estimated coefficients of these selected models are detailed in Table 6.
The first thing we observe in Table 5 and Table 6 is that the time trend coefficient is required in various cases: Argentina, Brazil, China, Egypt, Ethiopia, India, Russia, Saudi Arabia, South Africa and UAE, and, in all these countries, the parameter for the time trend exhibits a positive value, the highest one corresponding to Ethiopia (0.0499) followed by China (0.0384); for the remaining three countries (i.e. Brazil, Iran and Russia), the time trend coefficient is statistically insignificant. On the other hand, if we focus on the magnitude of the persistence by evaluating the estimated differencing parameter d, we find that reversion to the mean only takes place in two cases: Iran, with an estimate of d equal to 0.44, and Saudi Arabia (d = 0.66); in these two cases, the values of d in the 95% confidence intervals are all below 1, implying transitory shocks. In all the other cases, though the values of d are smaller than 1 in some cases, the interval includes the value 1, implying that the presence of a unit root cannot be ruled out. In the case of Ethiopia, the estimated value of d is 0.08, but the confidence interval is sufficiently broad that neither hypothesis can be rejected, the I(0) or the I(1) hypotheses.
Conclusion
We have investigated in this study the persistence in the series corresponding to consumption levels in the BRICS Plus nations. For this purpose, we have looked at data corresponding to spending by individuals and NPISHs in a set of 11 countries: Argentina, Brazil, China, Egypt, Ethiopia, India, Iran, Russia, Saudi Arabia, South Africa, and UAE.
The data periods span from 1960 to 2022, annually. The results reveal some differences across countries. Thus, positive time trends are detected in all countries except Brazil, Iran and Russia, and the highest trends are observed in Ethiopia and Saudi Arabia for households and NPISHs final consumption expenditure. More importantly, empirical support for mean-reverting behaviour and transitory innovations is exclusively detected in the instances of Iran and Saudi Arabia; for the remaining nations, the null hypothesis of permanency of shocks cannot be rejected.
The lack of significance trends in Brazil, Iran and Russia could be caused by economic factors (economic downturns, recessions or unfavourable economic policies), political instability (political uncertainties, corruption issues or policy changes), social factors (societal issues, demographic changes or cultural factors) and external shocks (events like natural hazards, global economic crises or geopolitical tensions).
Correlating this conclusion with certain aspects of the literature we have reviewed reveals that diverse topics yield varied conclusions. In our analysis, we consider the inclusion of new members, yet in comparison with the established ones (BRICS), Brazil and Russia appear to be at a disadvantage in terms of consumption.
Notably, according to the study by An et al. (2022), for the examination of external liquidity shocks, India, South Africa and Brazil are implicated, while for COP shocks, South Africa and Russia are the focal points (Su et al. 2021). Each country exhibits distinct strengths and weaknesses across different domains, as discerned through the various analyses. Other analyses in this vein are recommended to explore additional topics where the BRICS Plus countries may be sensitive to shocks.
The direct integration of our findings on fractional integration with the historical chronology of consumption theories provides a mathematical synthesis that validates Brown’s (1952) persistence of habit theory.
Consumption today and tomorrow depends on its past. Studying historical data allows us to make decisions today about what might happen tomorrow, thus eliminating uncertainty, enabling efficient decision-making and providing market measurement. By utilising a fractional integration framework, this study moves beyond the restrictive binary choice of treating consumption as either stationary (d = 0) or non-stationary (d = 1). Instead, analysing consumption across the spectrum of d-values allows for a deeper and clearer reconciliation of macroeconomic theory and policy. If empirical estimations yield a parameter within the stationary long-memory range (0 < d < 0.5), it provides empirical validation for a mild interpretation of Brown’s (1952) habit persistence. In this scenario, consumption displays behavioural memory, meaning fiscal policies will possess an extended transmission mechanism, trickling through aggregate demand over an elongated horizon rather than exhausting their impact immediately. Critically, if the data reveal a non-stationary but mean-reverting process (0.5 ≤ d < 1), the analytical interpretation shifts profoundly. This range indicates a severe structural inertia that aligns closely with Duesenberry’s (1948) ‘ratchet effect’ and an intense manifestation of Brown’s habit persistence. Households do not smooth consumption evenly as Hall (1978) suggests (d = 1); instead, they experience a prolonged, non-stationary drift, aggressively maintaining their peak living standards in the wake of negative shocks by depleting savings or accumulating debt. For policymakers, an economy characterised by 0.5 ≤ d < 1 warns that temporary stabilisation policies are fundamentally inadequate. Because the behavioural habits of consumers cause economic shocks to persist deeply and decay at a hyperbolic, glacially slow rate, any fiscal intervention must be structurally sustained. Temporary relief checks will merely be swallowed up by households trying to patch over their accumulated deficits to maintain their baseline habits, whereas sustained policy adjustments are required to safely guide the economy back to its long-run equilibrium mean. For details on the integration of theories, findings and policies, see Table 7.
| TABLE 7: Summary of the integration of theories, findings and policies. |
For countries experiencing a lack of time trends and in particular those with evidence of permanent shocks, it is crucial for authorities to implement robust measures aimed at restoring the initial consumption patterns. The causes of reduced consumption can be diverse, necessitating a comprehensive approach. Here are various suggestions and policy implications for enhancing consumption:
- Fiscal stimulus: Deploy expansionary fiscal policies such as tax cuts (value added tax for vital goods and services) or direct cash transfers to augment disposable income, fostering increased consumer spending (Kaplan & Violante 2014; Lyssiotou & Savva 2021).
- Enhancing disposable income: Pursue policies focusing on wage improvement, employment boosting measures and job creation to elevate household income, resulting in heightened consumption levels (Banker, Byzalov & Chen 2013; Yasar 2017).
- Infrastructure investment: Invest in infrastructure projects to generate jobs, improve transportation networks and stimulate overall economic activity, consequently impacting consumption positively (Ramey 2020).
- Access to credit: Facilitate access to credit for households, enabling them to make more significant purchases and investments, thereby stimulating consumption (Kus 2013).
- Education and financial literacy: Promote financial literacy programmes and educational initiatives to empower individuals in making informed financial decisions, leading to improved consumption patterns (Dinkova, Kalwij & Alessie 2021).
- Favourable business environment: Implement policies supporting entrepreneurship, innovation and a conducive business environment, fostering economic growth, job creation and increased consumer spending (Wüstenhagen et al. 2008).
- Enhancing international trade: Expand international trade to increase market access, encourage competition and provide consumers with a broader range of goods and services at competitive prices, potentially stimulating consumption (Janeba 2007).
These recommendations are offered in response to exogenous shocks, considering the elevated magnitude of persistence observed by the time series across the entire sample of nations. Finally, other approaches can be implemented to corroborate the results reported in this work. Thus, for example, other economic variables may be examined accompanying consumption in the interrelationships among variables, utilising, for instance, the fractional cointegrated vector autoregressive (FCVAR) framework developed by Johansen and Nielsen (2010, 2012), or the more recent panel fractionally integrated approach of Ergemen and Velasco (2017) might be alternative viable approaches. Research along these lines is currently underway.
Acknowledgements
An internal project from the Universidad Francisco de Vitoria is acknowledged, as well as the Commerce and Tourism Faculty of the Universidad Complutense de Madrid for their support.
This article includes visual content (Figure 1) generated with the assistance of OpenAI ChatGPT, 2026. The authors reviewed, modified, and validated all AI-generated artefacts and take full responsibility for their accuracy and integrity.
Competing interests
The authors reported that they received funding from MCIN/AEI/10.13039/501100011033 from ‘Ministerio de Ciencia e Innovación’ (MICIN), ‘Agencia Estatal de Investigación’ (AEI) Spain and ‘Fondo Europeo de Desarrollo Regional’ (FEDER) which may be affected by the research reported in the enclosed publication. The authors have disclosed those interests fully and have implemented an approved plan for managing any potential conflicts arising from their involvement. The terms of these funding arrangements have been reviewed and approved by the affiliated University in accordance with its policy on objectivity in research.
CRediT authorship contribution
Luis A. Gil-Alana: Data curation, Formal analysis, Funding acquisition, Methodology, Validation, Writing – original draft, Writing – review & editing. Gema Lopez: Conceptualisation, Data curation, Investigation, Writing – original draft, Writing – review & editing. Both authors reviewed the article, contributed to the discussion of results, approved the final version for submission and publication and take responsibility for the integrity of its findings.
Funding information
This work was supported by the MCIN/AEI/10.13039/501100011033 from ‘Ministerio de Ciencia e Innovación’ (MICIN), ‘Agencia Estatal de Investigación’ (AEI) Spain and ‘Fondo Europeo de Desarrollo Regional’ (FEDER) under grant PID2023-149516NB-I00.
Data availability
Data are available upon request from the corresponding author, Gema Lopez. Additionally, this public data related to Households and NPISHs Final consumption expenditure can be found at: https://data.worldbank.org/indicator/NE.CON.PRVT.CD.
Disclaimer
The views and opinions expressed in this article are those of the authors and are the product of professional research. They do not necessarily reflect the official policy or position of any affiliated institution, funder, agency, or that of the publisher. The authors are responsible for this article’s results, findings, and content.
References
Al-Mohamad, S., Jreisat, A. & Sraieb, M., 2023, ‘An experimental analysis of how resilient BRICS stock markets are to the Ukrainian Crisis’, Available at SSRN 4329351. https://doi.org/10.2139/ssrn.4329351
An, H., Wang, H., Delpachitra, S., Cottrell, S. & Yu, X., 2022, ‘Early warning system for risk of external liquidity shock in BRICS countries’, Emerging Markets Review 51, 100878. https://doi.org/10.1016/j.ememar.2021.100878
Baillie, R.T., Chung, C.-F. & Tieslau, M.A., 1996, ‘Analysing inflation by the fractionally integrated ARFIMA–GARCH model’, Journal of Applied Econometrics 11(1), 23–40. https://doi.org/10.1002/(SICI)1099-1255(199601)11:1<23::AID-JAE374>3.0.CO;2-M
Banker, R.D., Byzalov, D. & Chen, L.T., 2013, ‘Employment protection legislation, adjustment costs and cross-country differences in cost behavior’, Journal of Accounting and Economics 55(1), 111–127. https://doi.org/10.1016/j.jacceco.2012.08.003
Barkoulas, J., Baum, C.F. & Oguz, G.S., 1996, ‘Fractional dynamics in a system of long term international interest rates’, Working Papers in Economics, 318, viewed 05 October 2023, from http://fmwww.bc.edu/EC-P/wp315.pdf.
Batondo, M. & Uwilingiye, J., 2022, ‘Comovement across BRICS and the US stock markets: A multitime scale wavelet analysis’, International Journal of Financial Studies 10(2), 27. https://doi.org/10.3390/ijfs10020027
Bhargava, A., 1986, ‘On the theory of testing for unit roots in observed time series’, The Review of Economic Studies 53(3), 369–384. https://doi.org/10.2307/2297634
Boulding, K.E., 1945, ‘The consumption concept in economic theory’, The American Economic Review 35(2), 1–14, viewed 22 May 2026, from https://www.jstor.org/stable/1818427.
Brown, T.M., 1952, ‘Habit persistence and lags in consumer behaviour’, Econometrica: Journal of the Econometric Society 20(3), 355–371. https://doi.org/10.2307/1907409
Browning, M. & Crossley, T.F., 2001, ‘The life-cycle model of consumption and saving’, Journal of Economic Perspectives 15(3), 3–22. https://doi.org/10.1257/jep.15.3.3
Caporale, G.M. & Gil-Alana, L.A., 2018, ‘Unemployment in Africa: A fractional integration approach’, South African Journal of Economics 86(1), 76–81. https://doi.org/10.1111/saje.12178
Cavallo, A., 2024, ‘Inflation with Covid consumption baskets’, IMF Economic Review 72(2), 902–917. https://doi.org/10.1057/s41308-023-00213-y
De Haan, J., Lundström, S. & Sturm, J.-E., 2006, ‘Market-oriented institutions and policies and economic growth: A critical survey’, Journal of Economic Surveys 20(2), 157–191. https://doi.org/10.1111/j.0950-0804.2006.00278.x
Diebold, F.X. & Rudebusch, G.D., 1989, ‘Long memory and persistence in aggregate output’, Journal of Monetary Economics 24(2), 189–209. https://doi.org/10.1016/0304-3932(89)90003-2
Diebold, F.X. & Rudebusch, G.D., 1991, ‘On the power of Dickey-Fuller tests against fractional alternatives’, Economics Letters 35(2), 155–160. https://doi.org/10.1016/0165-1765(91)90163-F
Diebold, F.X. & Yilmaz, K., 2012, ‘Better to give than to receive: Predictive directional measurement of volatility spillovers’, International Journal of Forecasting 28(1), 57–66. https://doi.org/10.1016/j.ijforecast.2011.02.006
Dickey, D.A. & Fuller, W.A., 1979, ‘Distributions of the estimators for autoregressive time series with a unit root’, Journal of American Statistical Association 74(366), 427–481. https://doi.org/10.1080/01621459.1979.10482531
Dinkova, M., Kalwij, A. & Alessie, R., 2021, ‘Know more, spend more? The impact of financial literacy on household consumption’, De Economist 169(4), 469–498. https://doi.org/10.1007/s10645-021-09391-4
Duesenberry, J.S., 1948, ‘Income-consumption relations and their implications’, in L. Metzler, P.A. Samuelson, A.P. Lerner & J.T. Dunlop (eds.), Income, employment and public policy, pp. 54–81, WW Norton & Company, Inc, New York, NY.
Elavarasan, R.M., Nadarajah, M., Pugazhendhi, R., Sinha, A., Gangatharan, S., Chiaramonti, D. et al., 2023, ‘The untold subtlety of energy consumption and its influence on policy drive towards Sustainable Development Goal 7’, Applied Energy 334, 120698. https://doi.org/10.1016/j.apenergy.2023.120698
Elliot, G., Rothenberg, T.J. & Stock, J.H., 1996, ‘Efficient tests for an autoregressive unit root’, Econometrica 64, 813–836. https://doi.org/10.3386/t0130
Ergemen, Y.E. & Velasco, C., 2017, ‘Estimation of fractionally integrated panels with fixed effects and cross-section dependence’, Journal of Econometrics 196(2), 248–258. https://doi.org/10.1016/j.jeconom.2016.05.020
Expansión, 2023, Datosmacro, Datosmacro.com, viewed 03 October 2023, from https://datosmacro.expansion.com/.
Friedman, M., 1957, A theory of the consumption function, a study by the national bureau of economic research, Princeton University Press, New York, NY.
Gil-Alana, L.A. & Robinson, P.M., 1997, ‘Testing of unit root and other nonstationary hypotheses in macroeconomic time series’, Journal of Econometrics 80(2), 241–268. https://doi.org/10.1016/s0304-4076(97)00038-9
Hall, R.E., 1978, ‘Stochastic implications of the life cycle-permanent income hypothesis: Theory and evidence’, Journal of Political Economy 86(6), 971–987. https://doi.org/10.1086/260724
Hassan, Q., Al-Hitmi, M., Tabar, V.S., Sameen, A.Z., Salman, H.M. & Jaszczur, M., 2023, ‘Middle east energy consumption and potential renewable sources: An overview’, Cleaner Engineering and Technology 12, 100599. https://doi.org/10.1016/j.clet.2023.100599
Hassler, U. & Wolters, J., 1994, ‘On the power of unit root tests against fractional alternatives’, Economics Letters 45(1), 1–5. https://doi.org/10.1016/0165-1765(94)90049-3
Hu, Y. & Man, Y., 2023, ‘Energy consumption and carbon emissions forecasting for industrial processes: Status, challenges and perspectives’, Renewable and Sustainable Energy Reviews 182, 113405. https://doi.org/10.1016/j.rser.2023.113405
Ibrahim, R.L., Ozturk, I., Al-Faryan, M.A.S. & Al-Mulali, U., 2022, ‘Exploring the nexuses of disintegrated energy consumption, structural change, and financial development on environmental sustainability in BRICS: Modulating roles of green innovations and regulatory quality’, Sustainable Energy Technologies and Assessments 53, 102529. https://doi.org/10.1016/j.seta.2022.102529
Ivanovich, C.C., Sun, T., Gordon, D.R. & Ocko, I.B., 2023, ‘Future warming from global food consumption’, Nature Climate Change 13(3), 297–302. https://doi.org/10.1038/s41558-023-01605-8
Janeba, E., 2007, ‘International trade and consumption network externalities’, European Economic Review 51(4), 781–803. https://doi.org/10.1016/j.euroecorev.2006.07.002
Johansen, S. & Nielsen, M.Ø., 2010, ‘Likelihood inference for a nonstationary fractional autoregressive model’, Journal of Econometrics 158(1), 51–66. https://doi.org/10.1016/j.jeconom.2010.03.006
Johansen, S. & Nielsen, M.Ø., 2012, ‘Likelihood inference for a fractionally cointegrated vector autoregressive model’, Econometrica 80(6), 2667–2732. https://doi.org/10.3982/ECTA9299
Jones, A., Nock, D., Samaras, C., Qiu, Y.L. & Xing, B., 2023, ‘Climate change impacts on future residential electricity consumption and energy burden: A case study in Phoenix, Arizona’, Energy Policy 183, 113811. https://doi.org/10.1016/j.enpol.2023.113811
Kaplan, G. & Violante, G.L., 2014, ‘A model of the consumption response to fiscal stimulus payments’, Econometrica 82(4), 1199–1239. https://doi.org/10.3982/ECTA10528
Kearney, J., 2010, ‘Food consumption trends and drivers’, Philosophical Transactions of the Royal Society B. Biological Sciences 365(1554), 2793–2807. https://doi.org/10.1098/rstb.2010.0149
Keynes, J.M., 1937, ‘The general theory of employment’, The Quarterly Journal of Economics 51(2), 209–223. https://doi.org/10.2307/1882087
Kus, B., 2013, ‘Credit, consumption, and debt: Comparative perspectives’, International Journal of Comparative Sociology 54(3), 183–186. https://doi.org/10.1177/0020715213504280
Kwiatkowski, D., Phillips, P.C.B., Schmidt, P. & Shin, Y., 1992, ‘Testing the null hypothesis of stationarity against the alternative of a unit root: How sure are we that economic time series have a unit root?’, Journal of Econometrics 54(1–3), 159–178. https://doi.org/10.1016/0304-4076(92)90104-Y
Lee, D. & Schmidt, P., 1996, ‘On the power of the KPSS test of stationarity against fractionally-integrated alternatives’, Journal of Econometrics 73(1), 285–302. https://doi.org/10.1016/0304-4076(95)01741-0
Lopez, G. & Gil-Alana, L.A., 2023, ‘Us vehicles sales. Evidence of persistence after COVID-19’, PLoS One 18(2), e0281906. https://doi.org/10.1371/journal.pone.0281906
Lyssiotou, P. & Savva, E., 2021, ‘Who pays taxes on basic foodstuffs? Evidence from broadening the VAT base’, International Tax and Public Finance 28(1), 212–247. https://doi.org/10.1007/s10797-020-09605-6
Muhammad, B., Khan, M.K., Khan, M.I. & Khan, S., 2021, ‘Impact of foreign direct investment, natural resources, renewable energy consumption, and economic growth on environmental degradation: Evidence from BRICS, developing, developed and global countries’, Environmental Science and Pollution Research 28(17), 21789–21798. https://doi.org/10.1007/s11356-020-12084-1
Nawaz, M.A., Hussain, M.S., Kamran, H.W., Ehsanullah, S., Maheen, R. & Shair, F., 2021, ‘Trilemma association of energy consumption, carbon emission, and economic growth of BRICS and OECD regions: Quantile regression estimation’, Environmental Science and Pollution Research 28(13), 16014–16028. https://doi.org/10.1007/s11356-020-11823-8
Ng, S. & Perron, P., 2001, ‘Lag length selection and the construction of unit root tests with good size and power’, Econometrica 69(6), 1519–1554. https://doi.org/10.1111/1468-0262.00256
O’neill, J., 2001, Building better global economic BRICs, vol. 66, pp. 1–16, Goldman Sachs, New York, NY, viewed 02 October 2023, from http://www.elcorreo.eu.org/IMG/pdf/Building_Better_Global_Economic_Brics.pdf.
OpenAI, 2026, ‘BRICS original members and BRICS plus additional members’ [AI-generated image], Generated using ChatGPT (image generation model) from a structured prompt, viewed n.d., from https://chatgpt.com.
Papier, K., Knuppel, A., Syam, N., Jebb, S.A. & Key, T.J., 2023, ‘Meat consumption and risk of ischemic heart disease: A systematic review and meta-analysis’, Critical Reviews in Food Science and Nutrition 63(3), 426–437. https://doi.org/10.1080/10408398.2021.1949575
Phillips, P.C.B. & Perron, P., 1988, ‘Testing for a unit root in time series regression’, Biometrika 75(2), 335–346. https://doi.org/10.1093/biomet/75.2.335
Ramey, V.A., 2020, The macroeconomic consequences of infrastructure investment (No. c14366), National Bureau of Economic Research, Chicago.
Reuters, 2023a, What is BRICS, which countries want to join and why? viewed 04 October 2023, from https://www.reuters.com/world/what-is-brics-who-are-its-members-2023-08-21/#:~:text=Over%2040%20countries%2C%20including%20Iran,2023%20summit%20chair%20South%20Africa.
Reuters, 2023b, BRICS invites six countries including Saudi Arabia, Iran to be new members, viewed 02 october 2023, from https://www.reuters.com/world/brics-invites-six-countries-including-saudi-arabia-iran-be-new-members-2023-08-24/.
Reuters, 2023c, Nuevo bloque BRICS no controlará 80% de la producción mundial de petróleo [New BRICS bloc will not control 80% of global oil production], viewed 04 October 2023, from https://www.reuters.com/fact-check/espanol/KPYDKEWZ5NKA3ENCEKA42HA2SM-2023-10-10/.
Robinson, P.M., 1994, ‘Efficient tests of nonstationary hypotheses’, Journal of the American Statistical Association 89(428), 1420–1437. https://doi.org/10.1080/01621459.1994.10476881
Rodrik, D., 2005, ‘Chapter 14 growth strategies’, Handbook of Economic Growth 1, 967–1014. https://doi.org/10.1016/S1574-0684(05)01014-2
Sachan, A., Sahu, U.K., Pradhan, A.K. & Thomas, R., 2023, ‘Examining the drivers of renewable energy consumption: Evidence from BRICS nations’, Renewable Energy 202, 1402–1411. https://doi.org/10.1016/j.renene.2022.11.080
Sadiq, M., Ou, J.P., Duong, K.D., Van, L. & Xuan Bui, T., 2023, ‘The influence of economic factors on the sustainable energy consumption: Evidence from China’, Economic Research-Ekonomska Istraživanja 36(1), 1751–1773. https://doi.org/10.1080/1331677X.2022.2093244
Samour, A., Adebayo, T.S., Agyekum, E.B., Khan, B. & Kamel, S., 2023, ‘Insights from BRICS-T economies on the impact of human capital and renewable electricity consumption on environmental quality’, Scientific Reports 13(1), 5245. https://doi.org/10.1038/s41598-023-32134-1
Schmidt, P. & Phillips P.C.B., 1992, ‘LM tests for a unit root in the presence of deterministic trends’, Oxford Bulletin of Economics and Statistics 54(3), 257–287. https://doi.org/10.1111/j.1468-0084.1992.tb00002.x
Stuenkel, O., 2020, The BRICS and the future of global order, Lexington Books, Lanham.
Su, C.-W., Huang, S.-W., Qin, M. & Umar, M., 2021, ‘Does crude oil price stimulate economic policy uncertainty in BRICS?’, Pacific-Basin Finance Journal 66, 101519. https://doi.org/10.1016/j.pacfin.2021.101519
The Business Standard, 2023, BRICS PLUS: China proposes new name as member states agree on expansion, viewed 02 October 2023, from https://www.tbsnews.net/world/brics-plus-china-proposes-new-name-member-states-agree-expansion-688274.
The World Bank, 2023a, GDP growth (annual %), viewed 05 October 2023, from https://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG.
The World Bank, 2023b, Households and NPISHs final consumption expenditure (current US$), viewed 11 November 2023, from https://data.worldbank.org/indicator/NE.CON.PRVT.CD.
U.S. Energy Information Administration, 2023, What countries are the top producers and consumers of oil? viewed 13 October 2023, from https://www.eia.gov/tools/faqs/faq.php?id=709&t=6.
Wang, Z., Bui, Q. & Zhang, B., 2020, ‘The relationship between biomass energy consumption and human development: Empirical evidence from BRICS countries’, Energy 194, 116906. https://doi.org/10.1016/j.energy.2020.116906
Warde, A., 2014, ‘After taste: Culture, consumption and theories of practice’, Journal of Consumer Culture 14(3), 279–303. https://doi.org/10.1177/1469540514547828
Wen, F., Shui, A., Cheng, Y. & Gong, X., 2022, ‘Monetary policy uncertainty and stock returns in G7 and BRICS countries: A quantile-on-quantile approach’, International Review of Economics & Finance 78, 457–482. https://doi.org/10.1016/j.iref.2021.12.015
Wüstenhagen, R., Sharma, S., Starik, M. & Wuebker, R., 2008, ‘Sustainability, innovation and entrepreneurship. New perspectives in research on corporate sustainability series: Introduction to the volume’, in Edward Elgar Limited (ed.), Sustainable innovation and entrepreneurship, pp. 1–23, Elgaronline, Cheltenham; Northampton, MA.
Yasar, N., 2017, ‘The relationship between energy consumption and economic growth: Evidence from different income country groups’, International Journal of Energy Economics and Policy 7(2), 86–97, viewed 22 November 2023, from https://dergipark.org.tr/en/pub/ijeeep/issue/31921/351183.
Yaya, O.S., Ogbonna, A.E., Mudida, R. & Abu, N., 2021, ‘Market efficiency and volatility persistence of cryptocurrency during pre- and post-crash periods of Bitcoin: Evidence based on fractional integration’, International Journal of Finance & Economics 26(1), 1318–1335. https://doi.org/10.1002/ijfe.1851
Yucel, A.G., Koksal, C., Acar, S. & Gil-Alana, L.A., 2022, ‘The impact of COVID-19 on Turkey’s tourism sector: Fresh evidence from the fractional integration approach’, Applied Economics 54(27), 3074–3087. https://doi.org/10.1080/00036846.2022.2047602
Zheng, M., Feng, G.-F., Zhao, X. & Chang, C.-P., 2023, ‘The transaction behavior of cryptocurrency and electricity consumption’, Financial Innovation 9(1), 44. https://doi.org/10.1186/s40854-023-00449-7
|
Hide
Show all
Email this article (Login required)
