Tiny house characteristics in Indonesia based on millennial’s user preference

ABSTRACT

Tiny house (TH) characteristics in Indonesia based on Millennial’s user preference (MUP) were presented. The characteristic of population, type of generation, and each city characteristic were first investigated for deciding MUP. Next, TH was investigated by analysing preferred housing sizes and prices based on MUP. Qualitative and quantitative analyses were conducted by an exploratory approach on textual and internet sources with a combination of quantitative online surveys with a statistical approach. As a result, the MUP was still firmly dominated by the stereotype of living in a big house with a large land area. In contrast, the realistic housing sizes and prices were narrowed down to a small house with a limited facilities compared to the stereotype-preferred house. Using Chi-square, TH that fits the MUP resulted in a combination of the price under 500 million Rupiah (approx. 33,695 USD) with a size of 60–80 m2. Preferably a below 80 m² property size, limited space in urban areas is able to be optimized as a settlement by applying the TH concept in Indonesia. By having such a house in urban areas, TH supports millennial’s modern cultural habits resulting in a high quality of living for productive generation of Indonesia.

KEYWORDS:

• Tiny house
• TH characteristics
• millennials
• user preferences
• housing affordability
• Indonesia

Introduction

Housing affordability is one of the main issues in many countries. The continued rise in large house prices causes social problems and indirectly affects the increase in ineffective land use. In many countries, large houses are generally limited to the wealthiest sectors of society, but in countries such as the United States of America (USA) and Australia, substantial houses are more commonplace among the middle classes (Shearer & Burton, 2019). On the other hand, these large and expensive houses are unsustainable in many ways; socially, economically, and environmentally. This issue was followed by the human settlement problem, such as limited land for housing continuously rising and emerging affordable housing needs.

Recently, tiny house (TH) promoted the comprehensive philosophy of small size, which is not a limitation but an opportunity to provide a beautiful, sustainable, affordable house. In developed countries such as the USA, the rise of TH significantly reduces the harmful environmental impact compared to large and conventional houses. The emerging TH, as a living place, is followed by the fundamental concept of ‘less is more’. TH preferences solve the limited availability of land due to the ineffective extensive house usage as a living place. Living in a TH enables people to afford a house in the city centre with a relatively lower budget since it only requires a smaller area of land (Becker, 2018). However, in the case of developing countries such as Indonesia, the user preferences for TH are determined from slightly different perspectives resulting in unsuitable practice compared to developed countries.

In big cities of Indonesia, complex issues related to urbanization and social perspectives dominate most Indonesians: a decent life is achieved by living in a big house with a large land area and facilities. This issue leads to rising demand for affordable houses in outskirt areas. By holding these perspectives, most Indonesians have to commute to their workplace daily, giving them other issues such as time-wasting due to traffic jams and increasing transportation expenses. On the other hand, Indonesians with an opposite perspective live in urban areas with limited affordable house availability forcing them to choose between government-subsidized houses or old houses, which are considered small houses (SH) with a basic, modest, and unruly design. Thus, Indonesians tend to live in SH due to having no other option due to the low purchasing power and economic condition (Anggraeni & Herlily, 2020). With these complexities, TH, with its superiority, is considered an upgrade for SH, leading to a potential solution to housing affordability issues in Indonesia.

In order to observe TH as a housing solution in Indonesia, composition of Indonesian and people generation who pays more attention to ongoing issues and positively responds toward new perspectives are considered. The demographic structure of Indonesia is dominated by the productive people, which shares about 68.7% (with about 39% proportion of millennial generation) of the total population in 2019 (National Development Planning Agency of Indonesia (Bappenas), 2021). This demographic composition benefits the Indonesian economy because a young population provides a large potential workforce that can accelerate economic growth. In the case of characteristics of people generation, the millennial generation has characteristics of being more concerned, giving more attention, and showing a positive attitude toward specific issues, for instance, saving the environment (Schmeltz, 2012). The millennial generation is highly influenced by their perceived behaviour which controls their house purchasing intention (Wijayaningtyas et al. 2019). Moreover, Indonesian millennial generation as the productive people facing the housing affordability issues due to the high price of housing in Indonesia. To some extent, housing prices on the land market, including rent and own house, are influenced by the city-level incomes. In addition, the rent fee in high-income cities such as the special capital region of Jakarta is more than three times that of low-income cities, such as the special region of Yogyakarta (Monkkonen, 2013). Therefore, we came up with the idea to characterize the tiny house for the Indonesian market, which is studied based on millennial user preferences. The millennial generation is considered a group of people who need a house the most. On the other hand, the high price of housing in Indonesia forces them to opt to purchase SH.

In this present study, three objectives are presented which are 1) to determine the MUP of housing needs in big cities of Indonesia, 2) to analyse the MUP of property prices and sizes as compared with personal income, and 3) to determine the MUP for realistic TH as a solution for housing needs in Indonesia. Finally, a discussion of the future prospect of TH in different aspects is presented to comprehend the study.

Research methodology

Figure 1 shows the flowchart of the study of TH characteristics in Indonesia based on MUP. An analysis of current status and challenges of housing in Indonesia is first conducted to deepen our understanding of the characteristics of population, type of generation, and city characteristics. Next, two methods of qualitative and quantitative analysis are conducted to characterize TH based on MUP. Here, the qualitative method focuses on the exploratory approach to textual and internet sources with a combination of quantitative online surveys with a statistical approach of chi-square. The study is designed to provide descriptive analysis regarding the recent housing problems and prospective tiny house (TH) market in Indonesia based on millennial user preference (MUP).

Figure 1. Flowchart of tiny house characteristics in Indonesia based on millennial’s user preference.

Housing in Indonesia: current status and challenges

Indonesia is an archipelago country with several big islands. Among them, Java Island is indisputable the most crowded island in Indonesia, which defines itself as the center of the economy where most Indonesians live. The population of five provinces in Java (which are the Special Capital Region of Jakarta, West Java, Central Java, Central Java, Special Region of Yogyakarta, and East Java) has relatively increased in the past couple of years. In these five provinces, the provincial capital city (which are Jakarta, Bandung, Semarang, Yogyakarta, and Surabaya) is considered the most representative area to characterize the city in each province. In addition, the Special Capital Region of Jakarta is categorized as the large city with more than one million residents (Monkkonen, 2013). In order to characterize each city, especially in relation to the housing shortage, several parameters related to population density, family size, dependency ratio, housing status, and sustainable development goals in housing projection are considered. Data are from the National Indonesian Socioeconomic Survey (SUSENAS) over several periods and modules (Statistics Indonesia, 2013b). Figure 2 shows the latest population of selected provinces in 2021. As the capital city of Indonesia, the Special Capital Region of Jakarta has the highest population density of 15,978 people per kilometer square (km²). Considering the area of 664.01 km², the Special Capital Region of Jakarta is the second smallest area, with the Special Region of Yogyakarta which possesses the smallest area of 32.5 km². Interestingly, the Special Region of Yogyakarta holds a relatively high population density of 1185 km² which is almost like the rest city with a wider area. For instance, Central Java with a lesser population density of 1120 km². In the period from 2013 to 2021, a slight increase in population density is observed in all provinces. In the past seven years, Central Java has had the highest rising by 5.70%, which is calculated from the population density in 2021 and its average population density. The lowest rise is observed in the Special Region of Yogyakarta by 0.5%.

Figure 2. Population density of selected provinces (a) in 2021, and (b) in period from 2013 to 2021.

Source: Author’s calculation with (Statistics Indonesia, 2013b)

In order to provide a comparable term, a normalized population density is introduced by dividing the population density by the highest value in the same time period (2013 to 2021). Figure 3 shows the normalized population density in each province. All provinces agree that the normalized population density linearly increases from 2013 to 2019. By 2020, only West Java and the Special Region of Yogyakarta had a slight decrement before a bounce back to rise again in 2021. These increments are followed by the slightly rising family population in all provinces, as shown in Figure 3. Figure 4 shows high number of family population in West Java, Central Java, and East Java. By considering both increments of overall and family population, the housing availability issue for inhabitants in each city is inevitable. In fact, house affordability has been a national issue that causes slum areas which decrease the quality of life in each city (National Development Planning Agency of Indonesia (Bappenas), 2021).

Figure 3. Normalized population density of selected provinces in period from 2013 to 2021.

Source: Author’s calculation with (Statistics Indonesia, 2013b)

Figure 4. Family population of selected provinces in period from 2013 to 2016.

Source: Author’s calculation with (Statistics Indonesia, 2013a)

The housing availability in Indonesia can be determined by analysing the housing status in each province. Two main statuses, which are rent and own house, are provided by Statistics Indonesia in the period from 2015 to 2021 (Statistics Indonesia, 2015a). Figure 5 shows the average proportion of housing status in each province. It is no doubt that the highest proportion of rented house is presented in Jakarta (also famous for its status as a metropolitan city in Indonesia) by 35.93%, which is almost the same as the proportion of own house by 48.30%. Even though the own house status in the rest of the provinces dominates the proportion by an average of >75%, still the housing affordability for specific generations needs to be investigated. In order to solve this problem, the dependency ratio is considered as the ratio of a number of people aged 0 to 14 and aged 65 and over, divided by the number of people aged 15 to 64 who are labelled as productive people. Figure 6 shows the dependency ratio of the selected province in the projection period from 2015 to 2035. As the ratio increases, the possibility of increasing the burden on the productive people of the population is high. This generation is responsible for maintaining the upbringing and adulthood of the economically dependent, as well as the main driver of the economic development of Indonesia.

Figure 5. Averaged proportion of housing status of selected province.

Source: Author’s calculation with (Statistics Indonesia, 2015a)

Figure 6. Dependency ratio of selected province in projection period from 2015 to 2035.

Source: Author’s calculation with (Statistics Indonesia, 2015b)

Qualitative TH perspective study

Qualitative TH perspective study aims to design a comprehensive online survey of MUP for TH in Indonesia. As the first step of this qualitative method, a literature review from several international journals and attending international conferences related to ethnographic of TH were conducted to deepen the understanding of TH. In the second step which is parallel to the first step, content analysis of popular social media, particularly TH groups on Instagram, was conducted. Visual media sources were analysed, including several YouTube series, such as ‘Living Big in A Tiny House’, which explored TH characteristics based on user. After finishing the first two steps, online surveys were designed using the platform of Google Forms Survey.

Data collection was obtained from two survey periods, as shown in Table 2, with two objectives which are 1) to explore the millennial’s user preference of TH in Indonesia, and 2) to deepen explore the millennials user preferences of TH in each city. The design of online surveys is shown in Figure 7(a). The surveys contain questions related to basic personal information, personal preference in house purchasing, and basic knowledge through the point of view to TH practice in Indonesia. Since the millennial respondents are Indonesians, the preferred language is Bahasa Indonesia. From the literature review and content analysis on Instagram TH groups, example housing designs as TH visualizations were selected and arranged for the online surveys (see Figure 7(b)). Two survey periods were conducted in June and December 2021 across the selected cities: Jakarta, Bandung, Semarang, Yogyakarta, Surabaya. Considering the city characteristics and normalized population density which are explained in previous section, the proportion of the sample in each city is set to be in the same amount to control the comparable term for the analysis. In this study, a threshold of 100 samples was chosen as the minimum sample size to get any kind of meaningful result (Singh & Masuku, 2018). Data collections were conducted by reaching the respondent of millennial through the social media groups such as WhatsApp society groups and randomly shared with different communities in selected cities. The millennial respondents were controlled by setting a range of age questions in the surveys. Hence, the evaluation of the surveys was continuously conducted during the period of data collections by checking the age ranges: 1981–1996 born, big cities coverage, and quantity of samples. After fulfilling all checklist requirements, the qualitative surveys data was analysed and quantified for investigating the objectives of this study. The status of online surveys is presented in Table 1 with two marks of ● and ○ which indicate the surveys are being implemented or not, respectively. In addition, the URL link of both online surveys are presented as follows,

Figure 7. (a) Opening page of online surveys delivered in Bahasa Indonesia, (b) Example of housing designs by @duastudio architecture and design studio (duastudio, 2018) for online surveys.

Table 1. URL link of online surveys in June and December 2021.

Period	Jakarta	Bandung	Semarang	Yogyakarta	Surabaya
Jun 2021	●	●	●	●	●
Dec 2021	●	○	●	●	○

Table 2. Chi-square critical value for $0.900>\alpha >0.010$ for r = 1–10.

	$P(X\le x)$
1 – α	0.100	0.900	0.950	0.975	0.990
r	$x_{\alpha =0.9}^2\left(r\right)$	$x_{\alpha =0.1}^2\left(r\right)$	$x_{\alpha =0.05}^2\left(r\right)$	$x_{\alpha =0.025}^2\left(r\right)$	$x_{\alpha =0.01}^2\left(r\right)$
1	0.016	2.706	3.841	5.024	6.635
2	0.211	1.605	5.991	7.378	9.210
3	0.584	6.251	7.815	9.348	11.34
4	1.064	7.779	9.488	11.14	13.28
5	1.610	9.236	11.07	12.83	15.09
6	2.204	10.64	12.59	14.45	16.81
7	2.833	12.02	14.07	16.01	18.48
8	3.490	13.36	15.51	17.54	20.09
9	4.168	14.68	16.92	19.02	21.67
10	4.865	15.99	18.31	20.48	23.21

June 2021: https://forms.gle/zCKqfQC9TjdGG7eg8

December 2021: https://forms.gle/M4qm1SQFWfyvW4jy7

Quantitative chi-square method

Quantitative Chi-square method aims to test the millennial respondent’s data independencies in big cities of Indonesia against the user preference parameters. In general, a chi-square distribution is a continuous distribution with degrees of freedom which is able to describe the distribution of a sum of squared random variables. In this study, the correlation between each MUP data was analysed by firstly determining three null hypotheses for each user preference parameters. Next, alternative hypotheses were decided as the opposite of the null hypotheses. The level of significance was set as α = 0.05 to show the strong evaluation against the null hypothesis. The millennial respondent’s data was calculated by the X² statistic as,

(1) $X^2=\sum _{i=1}^I\frac{{\left(H_0^k-n_i\right)}^2}{n_i}\dots \in {\mathbb{R}}^{I\times M}$(1)

where i is the number of user preference parameter, H₀ is the value of the null hypothesis, k is the variance of null hypothesis, n is the millennial respondent’s data under i-number, and M is the number of millennial respondent’s cities. In order to evaluate the significance of null hypotheses, the critical value based on degree of freedom r was calculated as,

(2) $r=\left(I-1\right)\times \left(M-1\right)$(2)

The reference value of X² under specific r refers to the standard chi-square critical value table as shown in Table 2.

Results

Statistic of millennial respondent

Quantitative analysis based on the online surveys is presented here. The surveys were distributed to 200 respondents with range age of millennial generation, 26–40 years old who have varied working backgrounds which are 34.5% as academics, 24.1% as employees, 20.7% as an entrepreneur, and other professions such as art performers, financial consultant, housewife, doctor, and freelancer. Figure 8 shows the distribution of millennial respondents age who took the surveys compared to its normal distribution. As compared with the normal distribution, the dominant age ranges are distributed from 30 to 36 years old, which has the same distribution of the millennial respondents who are 30–35 years old. In this age range, they are in their productive age, who already have income, a fixed career, and start married life. Since they start trying to settle down, they tend to look for a decent and comfortable house to live in for themselves and their families. Moreover, they tend to open their minds to options and offers to find functional and affordable houses. The other respondents, 26–29 years old, are in the phase of starting their careers. On the other hand, the age range of 36–40 years old respondents commonly already settled down. These two age ranges have a smaller proportion with a lower probability density function $\mathrm{P}<0.01$.

Figure 8. Distribution of respondents’ age.

Figure 9 shows the type of house based on MUP. Generally, the Millennials choose two types of residence as their place to live: landed houses and apartments (high-rise buildings). It was found that 95% of the respondents chose landed houses as their residence. Based on their point of view, they could have a better social life, more privacy, and ease of access by owning a landed house. The landed house has the potential to be expanded without many limitations and has a value of investment due to the clear ownership status of the land and building. Furthermore, the respondent’s cultural habits and norms culture that used to live in the landed house for years became the most crucial reason for choosing a landed house as a private residence.

Figure 9. Type of House based on MUP.

In order to investigate the most crucial reason for choosing a landed house, a parameter of the personal income of millennials was analysed. The income rate of respondents in each city varies depending on the regional minimum wage (RMW) applicable in each city. Among the five cities, the highest RMW is Jakarta at Rp4,453,935.00/month (approx. 300 USD), and the lowest is Yogyakarta at Rp2,153.970.00/month (approx. 145 USD). RMW of Semarang is Rp2,835,021,00/month (approx. 190 USD), Surabaya is Rp4,375,479,00/month (approx. 294 USD), and Bandung is Rp3.774.860,00/month (approx. 253 USD).

Figure 10 shows that the income range of approx. 67.33–336.67 USD is mostly owned by respondents who live in Yogyakarta. The income more than 674.33 USD is mostly owned by respondents who live in Jakarta, Semarang, and Bandung. The income range of 506.01–673.33 USD is mostly owned by respondents in Surabaya. Surabaya has same city characteristics as Jakarta as metropolitan and business city in Indonesia. These two cities have almost equal rate of RMW. On the other hand, Bandung and Yogyakarta have similarity as art and culture city. Meanwhile, Semarang has characteristics such as harbour and business city. Differences in city characteristics and RMW affect the MUP in house purchasing. Therefore, further data analysis is focused on three big cities: Jakarta, Semarang, and Yogyakarta.

Figure 10. Monthly Income of Respondents.

MUP of house purchasing

Property price based on personal income

Property price range is divided into four price ranges: under 500 million Rupiah (approx. 33,695 USD), 500–750 million Rupiah (33.695–50,500 USD), 750 million–1 billion Rupiah (50,500–67,400 USD), and more than 1 billion Rupiah (approx. 67,400 USD). These property price ranges are chosen based on the possible property prices that are affordable by the millennials with the amount of income as explained in the previous chapter (Figure 10). In the same data frame, property with price under 33,695 USD and 33.695–50,500 USD are favourable for respondents live in Yogyakarta. Meanwhile, respondents in Semarang and Jakarta are divided into three price ranges: under 33,695 USD, 33.695–50,500 USD, and 50,500–67,400 USD. Only a few of the respondents chose a property with price more than 67,400 USD (see Figure 11(a)).

Figure 11. (a) Distribution and (b) Trendline of property prices based on millennials income.

The trend of MUP on housing prices is shown in Figure 11(b). Yogyakarta has a high demand for property with a price under 33,695 USD due to the low rate of RMW. Property price in Yogyakarta is extraordinarily high and barely could not be afforded to middle-low class millennials. In the case of Semarang, the highest demand for property price is under 33,695 USD and gradually decreases for the higher property prices. Semarang has a slightly higher RMW as compared to Yogyakarta. However, property prices in Semarang are not as high as in Yogyakarta. Therefore, several millennials still consider buying property at higher prices. Meanwhile, Jakarta, the city with the highest RMW, has almost similar demand for property prices under 33,695 USD and 50,500–67,400 USD. Those considering buying a property with a price range of 50,500–67,400 USD are identified as millennials with an income of more than 674.33 USD. Jakarta has property prices which are almost similar to Yogyakarta. Even though the property prices are higher than in other cities, millennials in Jakarta are still able to afford them since they also have higher RMW.

Figure 12 shows the comparison of property mean price and preferred prices based on millennials income. By calculating the mean = 15 of each city as the threshold for determining the significance of property demand in each price, property price under 33,695 USD is preferred in Semarang and Yogyakarta. While in the case of property prices above 33,695 USD, the demand is available and commonly preferred but not as high as property prices under 33,695 US. In ideal economic conditions, approximately 60 m² house-size, identical to TH, could be purchased for under 33,695 USD. This condition shows that TH has the potential to accommodate the housing demand, which meets the purchasing power of buyers, in this case, millennials.

Figure 12. Comparison of property mean price and preferred prices based on millennials income.

Realistic property size based on personal income

Since the landed house is decided as one of the mandatory consideration factors by Millennials, the desired property size has a good agreement considering the property size they deemed to purchase within the available land area that is commonly available in an urban area. The property size range is divided into four price ranges: 40–60 m², 60–80 m², 80–100 m², and more than 100 m². These property size ranges are chosen based on the possible property size within the property price range explained in the previous section (see section 3.3.1). Also, these property sizes are commonly purchased in Indonesia. In the same data frame, Figure 13(a) shows millennials optimism about three cities’ capability to purchase a 60–80 m² property based on their income.

Figure 13. (a) Distribution and (b) Trendline of property sizes based on millennials income.

The trend of MUP on housing sizes is shown in Figure 13(b). The millennials in three cities believed they could still afford a 60–80 m² property size. The second options for property sizes that are possible to be purchased by millennials are 40–60 m² and 80–100 m². Based on Indonesia’s property market, property with a size of 40–60 m² is mainly possible to purchase with an average price under 33,695 USD. This property size is most identical to TH and typical property size that provided by government as subsidized house. However, most millennials tend to look for possibilities to get bigger property sizes that are still affordable. On the other hand, 80–100 m² property size is almost impossible to be purchased for less than 33,695 USD. Nevertheless, millennials might look for an opportunity to purchase an 80–100 m² property with specific terms and conditions based on their ability to afford it. In the case of millennials’ income, fewer millennials are optimistic about their capability to purchase big-size property, which is represented by more than 100 m² property size. Those who believe they have the capability to purchase big-size property are identified as respondents with an income of more than 674.33 USD.

Figure 14 shows the realistic property size based on millennials income. By calculating the mean = 18.75 of each city as the threshold for determining the significance of property demand in each size, a property size of 60–80 m² is favourable in all cities. In contrast, the demand for property sizes of 80–100 m² occurred in the two cities of Jakarta and Yogyakarta. Surprisingly, millennials in Semarang tend to keep their desire to purchase a property with a size of >100 m². These phenomena are considered as the effect of limited land availability in urban area. Overall, the demand for property smaller than 60–80 m² is not realistic for millennials. Based on these findings, a landed house with a property size of 60–80 m² is highly preferable among other options. This property size fits the TH, which has high flexibility to be designed and built as a 60–80 m² property size. In addition, the application of TH concept is needed to meet the millennials demand related to property size by developing compact space configuration and interior design. TH could be considered as the first phase of developing a living house.

Figure 14. Realistic property size based on millennials income.

Expected property size based on norm culture

The expected property size is investigated to investigate the original MUP. The expected property size is divided into four size ranges similar to the realistic property size explained in the previous section; those are 40–60 m², 60–80 m², 80–100 m², and more than 100 m². These property sizes are chosen based on property sizes that Indonesians commonly own. In the same data frame, Figure 15(a) shows the tendency of millennials in three cities observed to choose 80–100 m² and more than 100 m² property sizes as expected property sizes they have. This type of house is the typical property size owned by the previous generation, or in this case, are parents of millennials in Indonesia. However, the price of this house is believed to be above 33,695 USD in each city. The trend of data shown in Figure 15(b) confirms the big-size house as the expected property size.

Figure 15. (a) Distribution and (b) Trendline of expected property sizes based on millennials user preferences.

Figure 16 shows that more than 100 m² property size is favourable in all cities as the expected property size based on MUP. By calculating the mean = 155 of each city as the threshold for determining the significance of property demand in each size, a property size of 80–100 m² is favourable in all cities. Specifically, millennials in Semarang firmly choose more than 100 m² property size as expected property size. Its history as a trading city and port made Semarang dominated by traders in the past. Semarang became Dutch colonial headquarters during World War II as well. Those two events involved significant buildings such as housing and offices that transformed the mindset of Indonesians to have big houses. On the other hand, millennials in Yogyakarta confirm that more than 100 m² property size as expected and gradually decrease to the smaller property size. The topography influences this situation and the geographical condition of Yogyakarta as a small city, leading to adjusting the property size following the city morphology.

Figure 16. Expected property size based on millennials user preferences.

Meanwhile, millennials in Jakarta choose two ranges of property size as expected property size, which are 80–100 m² and more than 100 m² property size. The background as the centre of government and business city leads Jakarta to create a community that has the capability to purchase a big-size house, which seems to dominate the trend in the past. The difference in MUP between realistic and expected property sizes is contradictory. It is seen that each city’s norms culture and history affect the MUP regarding expected property size. In addition, the millennials were raised in a big-size house owned by their parents. This background built the mindset that a big size house is the ideal property size.

Discussions

Type of realistic house by MUP

Chi-square is used to validate the MUP correlations between each data independencies and to test the research hypothesis in order to determine criteria of tiny house that could accommodate millennials’ need to solve housing affordability. There are three variances $\mathrm{k}$ of null hypothesis ${\mathrm{H}}_0^{\mathrm{K}}$ which are:

1. absolute demand of property that is less than 33,695 USD = 100%;

2. demand of property that is less than 33,695 USD is gradually increased by 30% than the nearest property price range;

3. demand of property that is less than 33,695 USD is 30% higher than property with 50,500–67,400 USD price range and 80% higher than property with 50,500–67,400 price range.

By recalling Equations (1)(1) $X^2=\sum _{i=1}^I\frac{{\left(H_0^k-n_i\right)}^2}{n_i}\dots \in {\mathbb{R}}^{I\times M}$(1) and (2(2) $r=\left(I-1\right)\times \left(M-1\right)$(2) ), the Chi-square X² is calculated for each MUP parameters and the degree of freedom (df) is solved as $\mathrm{d}\mathrm{f}=(4-1)\times (3-1)=6$ which giving a critical value of 12.6 based on Table 1. The result of Chi-square analysis is shown in Table 3.

Table 3. Results of Chi-square analysis of property price based on millennials user preference.

X^2	Jakarta	Semarang	Jogja	Table
k = 1	5.067	2.8	17	12.6
k = 2	10.950	3.333	11.633	12.6
k = 3	38.307	15.428	4.114	12.6
${\mathbf{H}}_0^{\mathbf{k}}$
k = 1	TRUE	TRUE	FALSE
k = 2	TRUE	TRUE	TRUE
k = 3	FALSE	FALSE	TRUE

Based on Table 3, it is found that two cities have a similarity in level of property price demand. First is between Jakarta and Semarang, and second is between Semarang and Yogyakarta. This condition is influenced by the property price and MRW in each city. Even though the lowest property price is favorable, the demand for higher property price is still occurred since Jakarta has the highest MRW among all cities. With the lowest MRW among all cities, Yogyakarta has high demand for the lowest property price. In the case of Semarang, even though it does not have high MRW as Jakarta, millennials with an income more than 674.33 USD demanding for higher property prices. Based on chart as shown in Figure 17, the second hypothesis k = 2, which is ”demand of property that less than 33,695 USD is gradually increase by 30% than the nearest property price range”, shows the property price based on MUP is valid.

Figure 17. Property price based on millennials user preference for different variance of null hypotheses.

In the case of property size, three variances k of null hypothesis ${\mathrm{H}}_0^{\mathrm{k}}$ which are:

1. demand of 40–60 m² and 80–100 m² property size is equal, demand of 60–80 m² property size is 75% bigger than 40–60 m² and 80–100 m² property size;

2. demand of 60–80 m² property size is gradually 30% bigger than 40–60 m² property size and 100% bigger than 80–100 m² property size;

3. demand of 60–80 m² property size is 130% bigger than 40–60 m² and 80–100 m² property size and demand of more than 100 m² property size is 7% of the data population.

The result of Chi-square analysis is shown in Table 4 and Figure 18. The Chi-square analysis result of realistic property size based on MUP shows the similarity of property size demand in three cities observed. It is valid that 60–80 m² property size is the most favorable property size and 40–60 m² and 80–100 m² property size took the second priority whether the demand is equal or not. Meanwhile, more than 100 m² property size is not favorable due to the high price. In addition, the analysis of property price and realistic property size based on MUP emphasizes that millennials' income took big part in millennials' consideration of house purchasing.

Figure 18. Property size based on millennials user preference for different variance of null hypotheses.

Table 4. Results of Chi-square analysis of property size based on millennials user preference.

X^2	Jakarta	Semarang	Jogja	Table
k = 1	4.828	5.814	3.528	12.6
k = 2	7.495	6.674	10.095	12.6
k = 3	24.233	52.491	13.491	12.6
${\mathbf{H}}_0^{\mathbf{k}}$
k = 1	TRUE	TRUE	TRUE
k = 2	TRUE	TRUE	TRUE
k = 3	FALSE	FALSE	FALSE

TH Characteristics by MUP

In many countries, TH most likely formed as Tiny House on Wheels (THOW), a container house, or mobile TH in a caravan or RV (Ford & Gomez-Lanier, 2017; Weetman, 2018). Commonly, the size of TH is limited to 40 m² (Lau & Wei, 2018). However, the exact size of a TH is not defined yet, and the size of a TH varies on norms culture and the law applied in each country (Evans, 2019). In the case of Indonesia, we have applied the small house concept for a long time. Indonesian people used to live in small houses that were sufficient for their needs. However, this paradigm was shifted over time, leading to a new paradigm of a big house with a large yard. This type of house represents people’s achievement and could leverage their social image in society. As studied in the previous generation, the bigger population reflects the considerably bigger family size. It is common to have a lot of kids in a family. This situation leads them to have bigger house to accommodate the whole family. The bigger the house the more rooms are available in the house. However, millennials family tends to have less kids therefore they opt out a big size house. Conventional house in Indonesia commonly have a lot of rooms. The amount of the rooms is influenced by western house style which has many kinds of room, as shown in Figure 19. On the contrary, these rooms are low occupied, and the house price is significantly high.

Figure 19. Conventional house floor plan.

Source: Author’s design based on average conventional house floor plan.

Based on MUP, two findings related to the most essential room are defined: 1) several rooms such as guest room, storage, and backyard are not necessary; 2) several rooms can be integrated into a compact room without solid separation: living room, dining room, and kitchen (LDK). Figure 20 shows the TH floor plan based on MUP which shows that space configuration for millennials house could be more simplified. Effective space configuration for TH in Indonesia is interesting to be investigated.

Figure 20. Tiny house (TH) floor plan based on MUP.

Source: Author’s design based on average conventional house floor plan.

In TH, rooms that are necessary and highly occupied are categorized as private room and semi-private room. It means that everybody including guests who can come into the house are considered as part of family or close relatives. Meanwhile, common room is set as the room for meeting an ‘outsider’ which categorized as the outside of the house. TH is also a good chance for people who wants to make investment and provide an alternative option for people who still rent SH (considering the affordable rental rate).

In the case of the ideal house for the millennials, TH has become a promising solution due to its affordability and modern design, which helps them leverage their social image in society. By having such a house in urban areas, the millennials are able to commute easily to their workplace, amusement park, shopping mall, and other facilities supporting their modern cultural habits. TH prices and maintenance costs are priced lower than conventional landed houses and apartments. However, the limited selection of belongings that can be put in TH becomes one of the significant issues. For instance, compact and multi-function furniture is recommended to be included in the whole design of TH to maintain the layout effectiveness and functions (Shearer & Burton, 2019). Moreover, a built-in system related to electricity, gas, and water adds more initial cost to building a TH. These challenges are still able to be solved by considering the open plan concept, which integrates all rooms in the most effective ways and vertical extension potential (Nuraeny et al., 2020). TH in Indonesia is considered the first phase of developing a living house. The limited land area and compact property size lead to a vertical extension due to more rooms. In order to obtain this limited land area size, several unused areas are considered potential land areas, which leads to the solution of using the slum area in better ways.

In the case of the emerging TH in Indonesia, several issues can be discussed as follows. First, TH is commonly known as the small house, which is located in a slum area. In previous research, several cases of TH have been studied to solve the housing problem in slum areas. For instance, TH in Kampung Anak Kali at the banks of the Ciliwung River, Jakarta (Nuraeny et al., 2020) and TH in Kampung Muka, Jakarta (Anggraeni & Herlily, 2020). Both focused on how to bring up the solution for the low-income person regardless of their generation. However, as discussed in the previous section, TH itself can be defined as a significantly different type of house. In our study, we focused on how to bring up TH as the affordable housing which can be built in such an urban area and can be purchased by the most needed person, in this case, the millennial generation. The main motivation of our study is to promote and categorize a house that fits in such limited circumstances in big cities of Indonesia. Since the issue of land-use policy and housing development leads to a spike in housing prices (Monkkonen, 2013), TH can become the solution to housing affordability issues, especially to help the main driver of Indonesian economic growth is millennial generation. By considering their user preferences, TH can be adapted to any kind of limited area with the effective functionality of a living house. Moreover, TH can be the leader of a new, eco-friendly housing solution to combat the current waste of the housing industry (Ford & Gomez-Lanier, 2017). By taking account of the TH’s best practices in other countries, TH in Indonesia can be one of the accelerators for such a young community in the initial phase of their family life (Boeckermann et al., 2019). In addition, our study can contribute to the lack of academic literature surrounding this topic and to the emerging TH market in Indonesia.

TH community in Indonesia

The current situation shows that the millennial generation has started to consider TH as a house that suits them. Personal Instagram (IG) accounts that produce content related to TH practices in Indonesia are observed. Based on the information in the biography section, the account owners are identified as millennials and living in big cities such as Jakarta, Bekasi, and Bandung. Some of them stated that their house was a non-engineered house, which means that they designed the house by themselves and arranged the construction based on their knowledge. This finding is supported by the fact that the number of housing units built by private sector developers has more than 70 percent proportion of houses produced between 2002 and 2007 (Monkkonen, 2013). Commonly, they explain the reason for choosing TH as their house. Some of them have experience living in small apartments or houses in Japan. This experience enlightened them that living in a small and compact house makes their life simpler and teach them to organize their belonging much better. Also, some of them realized that TH is the only house size that is affordable based on their evaluation and projection of their financial condition. On that platform, they share tips and tricks about designing and living in a TH as well as updates related to household furniture and equipment that suits a TH.

The uniqueness of this phenomenon is that most of these accounts follow each other and transform into a virtual tiny house community in Indonesia. Each of them has more than 10,000 followers. It can be assumed that the followers are the millennial generation reminds that IG in Indonesia is used by the millennial generation and the younger generation. On each content they posted, it can be seen that they interact with each other in the comment section as well as on the other captions where the follower is able to ask questions related to how to initiate their own TH. The number of their followers leads producers or developers to collaborate with them in making a paid promotion or endorsement system to introduce household and furniture products. These producers are aware of the power of the TH community to promote their products as part of life simply in TH. This case shows that TH users have an important role in introducing the benefits of living in a TH to millennials, as their followers whose highly influenced by their perceived behaviour.

Conclusions

The millennial user preferences (MUP) for tiny house related to the prospective market in Indonesia have been successfully studied with conclusions as,

1. The MUP of housing needs in big cities of Indonesia is concluded to be a stereotype of the conventional point of view in which decent life is achieved by living in a big house with a large land area. In this regard, the millennials tend to choose a big house for living without any ability to purchase it.

2. The analysis of the MUP property price and realistic property size based on MUP emphasizes that millennial’s income took a prominent part in millennial’s consideration in house purchasing

3. The realistic property size is valid in 60–80 m² property size, which is the most favorable property size, and 40–60 m² and 80–100 m² property sizes took the second priority whether the demand is equal or not.

Based on the present study, TH is characterized as the compact house which fits to the limited circumstances in urban areas. By possessing the knowledge of TH, most millennials are expected to consider adopting the TH concept or purchase it for their living house. TH has become one of the promising solutions due to its affordability and modern design, which helps them to leverage their social image in society. By having such a house in urban areas, TH supports their modern cultural habits resulting in a high quality of living for productive generation of Indonesia.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

This work was supported by the Japan Science and Technology Agency [A22].