To Retire or Not to Retire? A Comprehensive Examination of Retirement Decision Dynamics in Italy

Abstract

The concept of active aging has become central to the public debate in many Western countries, given the increasing aging of the population and the future challenges associated with it. This phenomenon is tightly intertwined with choices regarding the retirement period, which can be postponed for different reasons by individuals in a society. The purpose of this paper is to understand the personal and family characteristics that influence future choices about retirement date in Italy. Utilizing data provided by the Bank of Italy in the Survey on Household Income and Wealth (SHIW), a logistic regression is performed to observe the impact of different covariates on the hypothesized retirement date of individuals close to retirement, i.e., over the age of 50 who are still in working status. The results show that the decision to delay retirement has increased in recent years, and is influenced by personal (e.g., matrimonial status), work, and economic factors. In addition, these factors have different impacts between males and females.

1. Introduction

Similar to other Western European countries and many other developed nations (i.e., Japan), Italy has encountered a notable upsurge in its population of individuals aged 65 years and above. Beyond the evident demographic, social, and healthcare implications, an aging populace presents a significant challenge to the labor market. On one hand, there will be a rising number of employed individuals over the age of 50, while the gradual increase in retirees might render the welfare system financially unsustainable for the government.

These two concerns are intricately connected by the pivotal decision of whether individuals should retire from the workforce upon reaching a certain age. This decision profoundly impacts both the labor force and subsequent economic policies, as well as the pension system. Historically, the employment rate for workers aged 55–64 exhibited two distinct phases: a decline until 1999, reaching 27.6%, and subsequent growth until 2022 [1]. Nonetheless, this rate still falls short of the 48.3% recorded across the European Union.

In this context, the concept of active aging becomes central. Active aging was defined by the World Health Organization (WHO) in 2002 as “the process of optimizing opportunities for health, participation, and safety to improve the quality of life of people as they age”. The concept of active aging has been on the agenda at the European level for several years, promoted through a multiplicity of documents and initiatives including the United Nations Madrid International Plan of Action on Aging (MIPAA), adopted by the Second World Assembly on Aging in Madrid in 2002; and the creation of the European Innovation Partnership on Active and Healthy Aging (EIPAHA) in 2011.The Italian labor market continues to exhibit a substantial and enduring gender disparity, which is a manifestation of its “patriarchal arrangement”. Nevertheless, in recent decades, there has been a deceleration in the traditional decline of female employment rates during the central and later phases of working life. However, significant variations persist in employment levels based on educational attainment, and a notable north–south divide remains prevalent.

The initial stage of the decline in the older workforce can be attributed to industrial restructuring, which disproportionately affected low-skilled workers, a demographic prominently represented within the 55–64 age group in Italy. To manage the surplus workforce, the government implemented considerable state interventions in the form of early-retirement policies, seniority pension schemes, and disability pensions. Over time, disability pensions emerged as the most widely utilized mechanism for regulating premature exits from the labor market [2].

The trajectory began to shift in the 1990s when a series of reforms were introduced with the aim of increasing the retirement age of the labor force. These reforms were implemented in an effort to address the prevailing trends and patterns in the Italian labor market.

Furthermore, a transition from an earnings-related pension system to a contribution-based model was implemented. The initial phase of development, spanning from World War II until approximately the 1980s, witnessed the provision of highly generous benefit formulas that primarily favored major labor market segments, notably, public employees and workers in large enterprises. In contrast, the safeguards offered to peripheral categories such as temporary and seasonal workers, the self-employed, and employees of small businesses were considerably more limited [3].

Starting from the mid-1970s, a significant process of transformation in the pension system unfolded in tandem with other changes impacting the economic–productive system, the labor market, and the demographic structure [4].

However, the most significant reforms were initiated at the outset of the 1990s in response to mounting financial pressures and apparent inequities in the distributive mechanism of the pension system. To facilitate the transition towards a multi-pillar system, pension policies introduced strategic interventions that simultaneously impacted the permanence of older workers in the labor market and augmented the range of services offered.

Broadly, over a period of 20 years (1992–2011), the reform process of the Italian pension system can be characterized by three alternating phases:

• An emergency phase spanning from 1992 to 1997, during which the “Amato reform” and the “Dini reform” were implemented to redefine the system and ensure the protection of old-age pensions.
• A parametric-type welfare reform enacted from the late 1990s until 2007, which aimed at recalibrating various parameters of the system.
• Subsequent emergency measures between 2008 and 2011 culminated in the Fornero reform, which is presently in effect and represents a pivotal component of the reform trajectory.

Regarding the retirement age, from 2019 the age for old-age retirement is set at 67 for all categories. In the ministerial decree of 5 November 2019, the age of 67 was confirmed until 2026, following ISTAT surveys that did not record an increase in life expectancy. In addition to the age requirement, there is a contribution requirement of 20 years: work contributions are valid to reach it. There is the possibility of early retirement by meeting certain requirements.

It can be accessed at any age; it is reserved for workers enrolled in INPS management schemes and requires:

-

42 years and 10 months of paid contributions for men (2227 weeks)

-

41 years and 10 months of paid contributions for women (2175 weeks)

For workers assigned to usurious or strenuous tasks, as defined by Legislative Decree No. 67 2011 and the 2018 Stability Law, it still remains in effect. For employees, the “Quota 97.6” requirement must be met, that is, the sum of the applicant’s age and years of contributions paid. For self-employed workers, “Quota 98.6” must be met. At least 35 years of paid contributions are needed in each case. In regard to women, it is a possibility reserved for female workers, introduced by Law 243 2004 on an experimental basis but extended until today, to retire with the contributory calculation system with 35 years of contributions (with the exclusion of notional contributions). From 2024, the age of 61 is required for all categories of workers with a discount of up to 2 years for those who have children, and membership in specific categories. Regarding incentives to postpone retirement, the Budget Law for 2023 provided a form of incentive to postpone retirement (Law 197/2022 Article 1 paragraphs 286/287).

The provision allows those who meet the requirements for a pension with quota 103 (62 years of age and 41 years of contributions) during 2023 to waive their dependent contributions at the time of postponing the time of access to retirement.

The corresponding amount is allocated by the employer directly in the employee’s paycheck. These amounts are subject to ordinary Irpef taxation.

In practice, delaying retirement makes it possible, on the one hand, to continue to receive wages, to which will be added the portion of contributions that no longer needs to be paid. Moreover, this will result in an increase in the pension once the worker retires from work.

In elucidating the determinants impacting the timing of labor-market withdrawal, the conceptual framework proposed by Hofacker and Radl [5] provides valuable insights by delineating three discrete yet interconnected levels: the macrolevel represented by welfare state configurations, encompassing the legislative modifications discussed earlier; the meso-level, characterized by workplace conditions; and the microlevel, focusing on individual attributes, which constitute the focal point of investigation in this study.

Concerning the microlevel, it is essential to recognize that retirement decisions may be subject to the influence of individual characteristics, including but not limited to educational attainment, health status, and personal relationships. These factors play a pertinent role in shaping an individual’s choice regarding labor market withdrawal and deserve careful examination in the context of the present research.

Education constitutes a crucial individual attribute significantly associated with the retirement age determination. The level of educational attainment stands as a principal determinant influencing a worker’s deliberation on whether to persist in the labor market or opt for withdrawal [6,7]. A higher educational level holds the potential to secure more remunerative employment opportunities, thereby offering increased prospects for career advancement and professional development.

Health status represents a vital individual attribute with a discernible impact on the determination of retirement age. Extensive research indicates that adverse health conditions, predominantly assessed through self-reported data, exhibit a robust association with early labor market exit. Within the context of Italy, Li Ranzi et al. [8] demonstrate a positive correlation between premature withdrawal from the workforce and chronic ailments, with gender differentials being noteworthy. Nevertheless, the influence of health on retirement decisions is not uniform across various income strata: individuals from lower socioeconomic backgrounds, despite facing health challenges, are compelled to remain in the labor force due to financial constraints, even when afflicted by illness.

Lastly, family attributes constitute another salient determinant in the decision-making process: the synchronization of retirement timing among married individuals [7], the distinctive features of the familial context in which they are embedded [9], and the role assumed by women within the family all exert substantial influence on the propensity to opt for early retirement or otherwise.

The primary objective of this research is to specifically investigate the timing of retirement [10], which includes the decision to continue working beyond that age, while also identifying the variables that may exert influence on such choices. Adopting a life-course approach offers a comprehensive and valuable perspective for comprehending the determinants of retirement decisions. This approach not only emphasizes the impact of conventional economic and health-related constraints but also considers a broad array of life-course determinants that contribute to the retirement transition [11].

Within the context of the legislative framework outlined earlier, the analysis will be primarily centered on individual characteristics previously mentioned, which are closely linked to personal attributes.

The research seeks to shed light, for Italy in the last two decades, on intentions for early or planned retirement under current legislation or whether individuals will retire in arrears. It analyzes, in fact, the intentions of individuals over the age of 50 on the timing of retirement (whether they plan, therefore, whether to retire on time or postpone) and the personal and family characteristics that may influence this decision.

This paper is organized as follows: Section 2 provides an overview of employment of older people in Italy in the last decade. Section 3 presents what factors may be associated with decisions on delayed or non-delayed retirement. Section 4 presents the data used in this research and the methodology applied. Section 5 presents the results obtained for Italy on retirement timing decisions and what factors influence it. Section 6 shows the conclusions.

2. Employment of the Older People in Italy

This section concentrates on the predicament of mature employees, aged between 50 and 74 years, within the Italian labor market throughout the past three decades, aiming to scrutinize the most salient transformations that have transpired. Commencing in the mid-1990s, subsequent to a precipitous downturn in the employment rate, the matter of older workers has garnered considerable prominence in the realm of Italian public discourse.

Over the past 30 years, the occupational status of senior Italian laborers has undergone two distinct phases, each spanning a comparable duration [2]. The initial phase, extending until 1999 (marked by the lowest recorded employment rate for older workers, standing at 27.6%), was characterized by a declining trajectory. However, a discernible shift emerged towards the conclusion of the 1990s, leading to a reversal of the preceding negative trend. In this subsequent phase, the employment rate for older workers exhibited a steady increase, culminating at just above 40% in 2022. This reversal in the employment rate can be attributed, to a certain extent, to a corresponding reversal in public policies designed to bolster employment and extend the retirement age for this demographic.

The manifestation of an imbalanced pension system grappling with unprecedented growth in social expenditure, along with elevated life expectancy and the progressive aging of the workforce, prompted a reassessment of the institutional approach that prevailed during the 1980s.

The growth of employment among individuals aged over 50 can be seen clearly in Figure 1 [1]. The observable time period in the graph spans from 2004 to 2022, which, as previously noted, falls within a phase of employment expansion. It is essential to observe, however, that the growth trend is continuous over the years and represents a significant improvement, with an increase of more than 10 percentage points.

Figure 2 presents a comprehensive overview of the employment trends among individuals aged 50–74 in different areas of Italy from 2018 to 2022. In 2018, the North and Center regions exhibited relatively similar employment rates, with 46.5% and 46.2%, respectively, while the South lagged at 35.0%. The overall Italian employment rate stood at 42.6%. The subsequent years witnessed a gradual increase in the employment rates across all regions. By 2022, the North and Center regions experienced a surge, reaching 48.1% and 48.0%, respectively, surpassing the South’s rate of 35.3%. The overall Italian employment rate reached 43.8%, suggesting a positive trend in the employment of individuals aged 50–74 across the country. These figures reflect dynamic regional variations, providing valuable insights into the evolving employment landscape for this demographic in Italy over the specified timeframe.

3. Life Course Factors That Affect Retirement Decisions

This research was centered on examining the influence of individual level factors on the timing of retirement. These individual life-course determinants pertain to personal circumstances of employees, encompassing aspects such as marital and parenthood statuses, educational attainment, gender, occupational status, and financial situation, which comprises state benefits and pension funds. The selection of these individual factors is rooted in the theoretical framework of the life-course schema, which typically distinguishes between two categories of determinants affecting career trajectories: life events in social contexts and positional factors [12,13].

Education stands as a significant individual attribute connected to the timing of retirement. The educational attainment of a worker constitutes a primary determinant that influences their choice between labor market participation and withdrawal. Empirical evidence reveals that individuals with higher educational levels tend to prolong their work tenure compared to those with lower educational attainment [6]. This phenomenon can be attributed to two key factors: first, the limited opportunities for requalification among the less educated workforce, and second, the delayed entry of highly qualified workers into the labor market, which results in a steeper income growth trajectory associated with age relative to other demographic categories [7]. The influence of education level on the decision to delay retirement is also found in other countries (e.g., Switzerland to the United States), where individuals with higher levels of education tend to delay retirement [14,15,16].

Health serves as another crucial individual characteristic that exerts an influence on the age of retirement. Various investigations have consistently demonstrated a robust link between suboptimal health conditions and early departure from the workforce. Notably, a study conducted by Alavinia and Burdorf [17] examines the correlation between self-perceived poor health and premature retirement across ten European countries. The findings indicate that, in seven out of these 10 countries, a positive and statistically significant association exists. For instance, in Italy, individuals experiencing poor health conditions face a substantial 45% higher risk of retirement, and furthermore, self-reported poor health strongly correlates with unemployment.

Li Ranzi et al. [8] conducted a comparable investigation focusing on Italy, utilizing data sourced from the Italian Health Interview Survey. Drawing upon data indicating a conspicuous discrepancy between the actual age of retirement and the legally mandated pension age, the researchers posited whether this phenomenon could be ascribed to a substandard health status or distinctive socioeconomic factors specific to the country, such as the notable generosity of the Italian public pension system. This study underscores that early retirement exhibits a significant positive correlation with chronic illness, with a notably higher risk observed among male individuals, individuals with lower educational attainment, and those belonging to the middle and working classes.

The likelihood of early retirement appears to be diminished among individuals of lower socioeconomic status, plausibly due to limited financial resources that may hinder their ability to cope with the income reduction entailed by retirement. In contrast, the nexus between chronic health issues and retirement exhibits heightened salience within the upper echelons of the social stratum: when an individual of elevated occupational social standing is afflicted by poor health, it exerts a pronounced positive influence on their decision to retire from the labor market [8].

Additionally, the synchronization of retirement between spouses significantly influences the timing of retirement decisions. A research study utilizing data from the Bank of Italy [7] examines two distinct models to assess the retirement trajectories of men and women. For men, one of the micro-factors associated with their retirement decision is the employment status of their partner [18]. The presence of an employed partner at the time of the man’s retirement positively influences the likelihood of his retirement. Conversely, if the partner is not employed, it exerts a negative impact on the probability of retirement, likely due to income-related considerations. Family size emerges as a significant factor, whereby larger families exhibit a decreased propensity for men to retire. This pattern is corroborated by a study involving both men and women conducted by Squarcio and Tuzi [19], which reveals that membership in a family with more than two members tends to delay one’s exit from the workforce, thus reducing the likelihood of early retirement.

Furthermore, retirement decisions manifest notable gender disparities. Women demonstrate a common inclination to postpone retirement, with female employees in Italy, residing in either central/northern or southern regions, tending to remain in the labor market for longer durations compared to men [7,9]. This phenomenon can be elucidated by both economic factors and personal motivations. Women’s careers tend to be more fragmented compared to those of male employees, often characterized by extended periods of temporary, low-paid employment before attaining stable positions. Some women have also exited the labor market due to maternity reasons and subsequently faced challenges in reentering after childbirth. All these factors compel them to prolong their tenure in the labor market to attain the requisite age limit or accumulate sufficient contributions to avail themselves of the pension system’s benefits [20].

In terms of intrinsic motivation, women exhibit a heightened level of interest in their professions, particularly among those with higher education and occupying managerial positions. A survey conducted by Abburrá and Donati [20] encompassing individuals aged 50–60 years, both employed and retired, residing in the Piedmont region of northwest Italy, reveals that a majority of the female respondents express a strong affection for their work, even when it is not correlated with high income levels. Indeed, beyond the two most prevalent reasons of fulfilling minimum pension requirements and augmenting income, the third prominent reason for women to persist in the workforce is their love for their job, with female respondents registering a seven-percentage point lead over their male counterparts. This trend is evident across all educational levels and occupational roles. In contrast, job attachment among men is characteristic primarily of those occupying high-ranking occupational positions, such as managers and executives.

Where, however, women are employed in care work or need to stay at home to care for relatives, studies show that they tend to withdraw from the workforce early [21,22]. This is especially the case in countries where public investment is limited in terms of caring for dependent persons or even the social environment tends to privilege one’s own family member for care instead of outside services.

Additionally, the survey highlights that a substantial proportion of employed women aged 50–60 express a willingness to postpone their departure from the labor market, including blue-collar workers, provided that more flexible and well-designed working hour arrangements, along with accessible care services, are made available to help balance family and work responsibilities. However, unlike many other European countries, the prevalence of part-time employment among older women in Italy is notably low, even lower than that observed among younger women. This disparity is particularly significant because caregiving responsibilities in Italy are predominantly borne by women, and older workers, in particular, face the dual commitment of caring for their own adult children with potential grandchildren and parents in need of assistance. Consequently, older female workers encounter considerable strain, viewing retirement as a strategy to cope with the burden of caregiving responsibilities at the expense of their aspirations for autonomy and personal fulfillment [20].

An investigation involving a sample of employees aged over 50, employed within a public organization in Italy, was conducted to explore various retirement intentions in relation to psychosocial factors [23]. The study revealed that experiencing learning difficulties and difficulties in adapting to change positively correlate with the inclination towards full retirement and negatively associate with the intention of job mobility, i.e., continuing to work in a different setting. Conversely, the perception of possessing adequate working skills encourages older workers to invest further in their career development. Similarly, the presence of job opportunities for growth and development is negatively linked to intentions of full retirement. Moreover, the fear of losing workplace relationships and experiencing a loss of social integration upon exiting the workforce enhances the desire to remain in the labor market or opt for part-time work.

Another key factor concerns the generosity of the pension system, both in economic terms and in terms of facilitating early exit from the labor force. This factor leads to substantial differences in the decision to delay retirement among many European countries. When pensions are not generous, workers are more likely to work longer [24,25]. Men retire unusually early in Belgium and France, for instance, which is explained by the generous retirement systems that provide relatively high replacement rates as well as a variety of early-retirement options within the framework [18,26]. Because of the financial incentives provided by national pension laws, widows and widowers, for instance, have a tendency to retire early in Germany but not in Spain [27].

In terms of employment position, self-employed individuals tend to retire later than employees, attributable to their higher statutory retirement age and less favorable rate of income replacement (i.e., the ratio between their final salary and pension) [7].

The choice of retirement exhibits a noticeable differentiation between physically demanding occupations and white-collar professions [6]. A study focusing on older workers in the Piedmont region [20] indicates that for blue-collar workers, the primary reason for retiring immediately upon fulfilling required contributions is a sense of exhaustion and weariness resulting from daily commuting and working hours (particularly for women), as well as a perception of outdated skills compared to younger generations (particularly for men). Conversely, among retirees with middle to high occupational roles, motives such as curiosity about “changing one’s life” and a desire for more time to pursue personal interests and hobbies are more prevalent.

4. Data and Methodology

The Survey on Household Income and Wealth (SHIW), a representative survey of Italian residents carried out by the Bank of Italy since the middle of the 1960s to gather data on income, saving, consumption expenditure, wealth, demographics, and labor force participation of Italian households, served as the foundation for this study¹.

Except for 1985, the SHIW was conducted annually until 1987, and then every two years until 2016 (the 1997 survey round was delayed to 1998). The edition of the SHIW for the year 2020, which the Bank of Italy finished in late 2021, was originally scheduled to be done in 2020 with reference to 2019 but was ultimately delayed to 2021 because to the COVID-19 epidemic.

The variable of interest² in the research, namely the prediction about the timing of retirement, was derived from a specific question asked of the individuals surveyed, who are asked at what age they plan to retire: if the answer reports a year equal to or less than the limit imposed by Italian legislation, it is considered early or on time retirement; if the answer reports a higher age, it is considered delayed retirement. In the case, however, where the respondent is already retired, the answer is about the age at which he or she retired. Thus, in this case, whether or not one retires is factual and does not concern a future intention of the respondent. This construction was necessary because there is no specific question about this in the questionnaire. The sample is represented only by subjects aged 50 years and older. This choice is justified by the fact that decisions about the timing of retirement will be made more properly as the statutory retirement age approaches. The age range for the total population varies from 50 to 89 years, with an average age of 56.11 years. Distinguishing by sex, the age range remains the same, but for women, the average age is slightly lower than for men (55.37 years compared to 56.51 for men).

The covariates used in the analysis concern several aspects that may influence the individual’s decision: first of all, a time component is included to observe how the subjects’ intentions change over the two decades under consideration with regard to personal and family characteristics, such as gender and marital status; the N. Household component refers to the number of household members, their area of residence, and their employment status and level of education; and economic variables include factors such as income class (divided into five quintiles) and whether or not the respondent is on state benefits (sickness benefits, redundancy payments, or other benefits from state or private agencies).

Table 1. Summary statistics.

Variable	Total	Male	Female
Retirement timing
Early and on-time retirement	66.89%	74.91%	50.09%
Late retirement	33.11%	25.09%	49.91%
Year
2000	9.53%	73.21%	26.79%
2002	9.60%	69.63%	30.37%
2004	10.18%	66.77%	33.23%
2006	10.95%	66.46%	33.54%
2008	8.35%	72.08%	27.92%
2010	8.53%	67.94%	32.06%
2012	9.27%	67.03%	32.97%
2014	10.09%	66.9%	33.1%
2016	11.64%	65.32%	34.68%
2020	11.86%	63.89%	36.11%
Sex
Male	67.69%
Female	32.31%
N° Household component (mean)	2.99	3.14	2.72
Marital status
Married	85.32%	91.56%	72.24%
Single	4.78%	3.81%	6.81%
Divorced	6.72%	3.5%	13.45%
Widower	3.19%	1.13%	7.5%
Area of residence
North	45.41%	43.93%	48.51%
Centre	20.82%	19.79%	22.96%
South and Islands	33.77%	36.27%	28.53%
Sector of employment
Agriculture	6.25%	6.19%	6.36%
Industry and construction	26.12%	32.74%	12.27%
Commerce	14.04%	12.85%	16.54%
Transportation and communication	4.69%	6.41%	1.58%
Finance	3.44%	3.99%	2.29%
Public administration	45.30%	37.82%	60.96%
Work status
Employee	74.19%	70.66%	81.91%
Self-employed	25.80%	29.44%	18.09%
Level of education
At most primary school	1.94%	1.98%	1.87%
Primary school	12.34%	13.77%	12.44%
Lower Secondary school	38.19%	38.96%	34.49%
Upper Secondary school	32.1%	30.93%	33.56%
Bachelor’s degree	14.41%	13.13%	17.1%
Post-graduate specialization	1.01%	1.23%	0.55%
Subsidies
No	96.09%	96.16%	95.93%
Yes	3.91%	3.84%	4.07%
Total observation	18,291	6444	11,846

Source: Survey on Household Income and Wealth (SHIW)—Bank of Italy.

presents the row statistics for the variables considered in the research.

Weighted logistic regression models are used to analyze the relationship between the covariates (a set of individual and household characteristics) and the likelihood of retiring early (i.e., early and ‘on-time’ retirement) or late. This model may be used to assess the effects of different factors on a discrete variable.

$$logit\left(p_{it}\right)={\beta }_t{x}_{it}+{\epsilon }_{it}.$$

(1)

The model calculates the probability that the dependent variable acquires value 1

$$E\left[Y_{ith}=1|X=x\right]=P\left(Y_{ith}=1\right),$$

(2)

where $P\left(Y_{ith}=1\right)$ represents the probability of observing the condition of success for the i-th individual given a particular value of X.

5. Results

Determinants of the choice of possible delayed retirement are observed in this session. In

Table 2. Logistic regression models of retirement timing: Italy (dependent variable: 1: late retirement, 0: early and on-time retirement.).

Covariates	Total	Female	Male
Year (reference = 2000)
2002	0.19	0.00	0.38
2004	0.19	0.20	0.10
2006	0.42 ***	0.13	0.71 ***
2008	0.60 ***	0.42 *	0.76 ***
2010	1.02 ***	0.84 ***	1.17 ***
2012	2.10 ***	1.85 ***	2.31 ***
2014	2.53 ***	2.14 ***	2.81 ***
2016	2.73 ***	2.48 ***	2.96 ***
2020	2.37 ***	2.24 ***	2.53 ***
Sex (reference = Male)
Female	1.32 ***
N° Household component	−0.04	0.05	−0.08 *
Marital Status (reference = Married)
Single	0.20	0.35	0.05
Divorced	0.20 *	0.39 ***	−0.24
Widower	0.28	0.40 *	−0.27
Area of residence (reference = North)
Centre	0.20 **	0.16	0.23 *
South and Islands	0.46 ***	0.52 ***	0.49 ***
Sector of employment (Reference = Agriculture)
Industry and construction	−0.05	0.45	−0.31
Commerce	0.22	0.69 ***	−0.01
Transportation and communication	−0.05	0.69	−0.34
Finance	−0.09	0.01	−0.29
Public administration	0.22	0.81 ***	−0.15
Work status (Reference = Employee)
Self-employed	0.91 ***	0.80 ***	0.93 ***
Level of education (reference = At most primary school)
Primary school	0.08	0.16	0.05
Lower secondary school	−0.16	−0.05	−0.20
Upper secondary school	0.21	0.27	0.17
Bachelor’s degree	0.76 *	0.52	0.90
Post-graduate specialization	1.68 ***	1.37	1.72 ***
Class of income (reference = 1st quintile)
2nd quintile	−0.30 *	−0.19	−0.33
3rd quintile	−0.63 ***	−0.74 ***	−0.58 ***
4th quintile	−0.68 ***	−0.80 ***	−0.67 ***
5th quintile	−0.76 ***	−1.13 ***	−0.56 ***
Subsidies (reference = No)		−0.24	0.12 ***
Yes	−0.02 ***
Constant	−3.93 ***	−1.72 ***	−2.45 ***
Observations	18,291	6444	11,846

Significant odds ratio (*** p < 0.01, ** p < 0.05, * p < 0.1).

we find the results for Italy. The variables analyzed have a very clear impact on decisions about the retirement period. First, it is interesting how, as the years passed, the willingness to continue working and delay retirement increased. Given the increasing post-2008 impact, this trend may also be due to uncertainty related to the national economic picture: uncertain economic conditions with worsening trends may lead individuals to decide to delay their retirement, concerned that their pension will not be sufficient to guarantee their previous living standards.

Family conditions also have a clear impact on retirement timing decisions and confirm the picture seen earlier in the literature. First of all, women confirm their intention to delay their exit from the labor force, either for personal or economic reasons [7,9]. Even with respect to the other personal characteristics considered, the economic aspect seems to be central to this choice: first of all, this consideration is evident if one analyzes income quintiles, where subjects with higher incomes tend to withdraw from the world of work according to the timeframe stipulated by the regulations.

Regarding the employment situation, it is interesting to note that the sector of work does not have a significant impact on the choice; what is important is the distinction between employees and the self-employed: the latter group, in fact, tend to consider retirement with more delayed timing than the former; again, this is probably for economic reasons and the possibility of continuing work independently. Furthermore, it is true—as previous research has also shown—that those with higher levels of education, such as those in professional or scientific occupations, exhibit greater job satisfaction, experience less health deterioration, and are more likely to continue working after the legal retirement age [16].

The second and third columns of the table present the results on the factors connected with retirement timing for both females and males. Retirement timing decisions may differ between men and women due to several factors noted earlier [20]. Noteworthy patterns emerge over the years, with consistent positive coefficients with the passing of the years in both female and male models, suggesting a steady increase in the likelihood of late retirement compared to the reference year 2000. The effect is more pronounced for males, with coefficients reaching 2.53 for the year 2020. Marital status exhibits diverse impacts; being single or divorced tends to increase the odds of late retirement for females, while, for males, marital status has no effect. Region-wise, residing in South and Islands is associated with higher odds of late retirement for both genders. Employment sector effects differ, with self-employed individuals consistently displaying higher odds of late retirement compared to employees. Education and income quintiles also play a role, indicating complex relationships between socio-economic factors and retirement timing. The higher level of education has a positive impact (increases the probability of delaying retirement) only for men, while a higher level of income has an opposite and similar impact for both sexes. The models provide a comprehensive understanding of the multifaceted determinants of retirement timing, emphasizing gender-specific nuances in these associations. The large number of observations enhances the robustness of these findings across the studied population.

6. Conclusions

The progressive aging of the population and the ever-increasing number of employed people of advanced age is presenting the Western world and Italy with new challenges on the labor market front and the sustainability of pension systems. Viewpoints on whether or not citizens need to lengthen their working lives so as not to burden an already ailing pension system, on the one hand, and the need to provide jobs for young people entering the labor market, on the other, make decisions about the timing of retirement very important today. Leaving aside macro-level aspects, such as the legislation currently applied in the country and the economic situation being experienced, although fundamental and certainly conditioning aspects, the research focuses on identifying the family, social, economic, labor, and geographical conditions that may influence the future choice of individuals with respect to retirement timing. In a nutshell: what are THE factors that lead individuals to decide to delay retirement?

The results seem to be quite clear: family and personal conditions crucial to the economic choice of continuing to work; these include having to maintain a larger household or, in the case of single individuals, the need to maintain an adequate level of income without the support of a partner [7]. Confirmation of the above comes from the analysis of income level, which shows us that as income level increases, people tend to retire earlier than subjects with lower income levels.

A separate and certainly more in-depth observation is deserved regarding the women’s issue: as observed in previous research [2,19], it is confirmed that women tend to remain in the world of work even after reaching retirement age; this is both for economic reasons, given women are typically paid less than men, interruptions due to pregnancies, and also a greater attachment to their work. The literature illustrated above, and the results of this research, emphasize the need to apply a range of good practices [19], given the increasing number of older workers and how they cope with the last years of their working lives, to improve the working conditions of those who wish to continue beyond the expected retirement age. These suggestions for interventions are key to making aging truly active by improving working and non-working conditions with targeted interventions. Starting with the central question of this paper, namely whether or not individuals intend to retire within the stipulated time frame, one possibility that can be offered to these workers is to ensure a range of flexible options, such as partial withdrawals from work or the possibility of scaling up work through part-time contracts. Related to this is the issue of redesigning the work tasks of older individuals, especially in the case of more strenuous jobs, ensuring that these individuals can cope with work in a healthy and risk-free manner. Finally, fundamental is the guarantee of continuous training for older workers: they are often only marginally involved in the processes of staff upgrading and training, but this would lead to a definite improvement in the working condition of the individual, who will feel further involved in the work processes.

Further investigations are, nevertheless, needed to understand the effects of meso- and macro-life-course determinants of the timing of retirement as well as of the voluntariness of retirement transitions and whether the reasons for retirement can be related to a voluntary delay in retirement or personal motivations such as job satisfaction or the possibility of proceeding in one’s career, or whether instead the decisions to retire postponed are “involuntary”, linked to a strictly economic motive of needing to continue with working life in order to be able to support oneself and one’s family.