Report: House Shopping in Australia
Project 2
SID 500497375
University of Sydney | MATH1005 | November 2020
1 Executive Summary
The main aim of this report is to outline the trends regarding home ownership and rent in Australia. The report also attempts to find out the average monthly mortgage in Australia and New South Wales and checks to see whether owning a home in New South Wales is more expensive than the rest of Australia (assuming you have a mortgage).
The main discoveries found in the report was that due to the high focus in Australia’s eastern corridor(New South Wales, Victoria and Queensland), these three states lead Australia in a majority of the data explored such as population, amount of people owning property and renting, however regarding the median monthly mortgage Western Australia took Queensland’s place. Tasmania and Victoria were found to have the highest population proportion regarding home ownership, with the Northern Territory leading in rent but being last in home ownership. Regarding New South Wales’ and Australia’s average monthly mortgage, New South Wales was found to be more expensive on average than the rest of Australia, with a difference of $215.42 per month.
2 Full Report
2.1 Initial Data Analysis (IDA)
# Load the data
houses = read.csv("C:/Users/georg/Documents/housing.csv", header = TRUE)
head(houses)## Local.Government.Area Owned...outright Owned...with.a.mortgage
## 1 LGA10050 Albury (C) 5440 6403
## 2 LGA10110 Armidale Dumaresq (A) 2888 2322
## 3 LGA10150 Ashfield (A) 4408 4240
## 4 LGA10200 Auburn (C) 5505 7104
## 5 LGA10250 Ballina (A) 6274 4169
## 6 LGA10300 Balranald (A) 327 216
## Rented...Real.estate.agent Rented...State.or.territory.housing.authority
## 1 3783 827
## 2 1803 451
## 3 4624 231
## 4 5824 836
## 5 2415 602
## 6 44 21
## Rented...Person.not.in.same.household
## 1 1067
## 2 519
## 3 972
## 4 1432
## 5 1065
## 6 86
## Rented...Housing.co.operative..community.or.church.group
## 1 144
## 2 103
## 3 158
## 4 130
## 5 135
## 6 20
## Rented...Other.landlord.type Rented...Landlord.type.not.stated
## 1 170 88
## 2 123 76
## 3 93 62
## 4 236 128
## 5 241 93
## 6 47 23
## Other.tenure.type Tenure.type.not.stated Total
## 1 149 545 18616
## 2 71 233 8589
## 3 125 384 15297
## 4 135 751 22081
## 5 199 404 15597
## 6 11 32 827
## Median.mortgage.repayment....monthly. State X X.1 X.2 X.3
## 1 1450 New South Wales NA NA NA NA
## 2 1430 New South Wales NA NA NA NA
## 3 2167 New South Wales NA NA NA NA
## 4 2000 New South Wales NA NA NA NA
## 5 1733 New South Wales NA NA NA NA
## 6 867 New South Wales NA NA NA NAnames(houses)## [1] "Local.Government.Area"
## [2] "Owned...outright"
## [3] "Owned...with.a.mortgage"
## [4] "Rented...Real.estate.agent"
## [5] "Rented...State.or.territory.housing.authority"
## [6] "Rented...Person.not.in.same.household"
## [7] "Rented...Housing.co.operative..community.or.church.group"
## [8] "Rented...Other.landlord.type"
## [9] "Rented...Landlord.type.not.stated"
## [10] "Other.tenure.type"
## [11] "Tenure.type.not.stated"
## [12] "Total"
## [13] "Median.mortgage.repayment....monthly."
## [14] "State"
## [15] "X"
## [16] "X.1"
## [17] "X.2"
## [18] "X.3"2.1.1 Size of Data
The size of the overall dataset is shown as: 559 obs. of 14 variables.
## Size of data
dim(houses)## [1] 559 182.1.2 R’s Classification of Data
Initially the database had many of the numerical values characterised as characters which would have not processed when creating graphs in the research questions. The csv file has since been modified to change all numerical values to integers.
class(houses)## [1] "data.frame"sapply(houses, class)## Local.Government.Area
## "character"
## Owned...outright
## "integer"
## Owned...with.a.mortgage
## "integer"
## Rented...Real.estate.agent
## "integer"
## Rented...State.or.territory.housing.authority
## "integer"
## Rented...Person.not.in.same.household
## "integer"
## Rented...Housing.co.operative..community.or.church.group
## "integer"
## Rented...Other.landlord.type
## "integer"
## Rented...Landlord.type.not.stated
## "integer"
## Other.tenure.type
## "integer"
## Tenure.type.not.stated
## "integer"
## Total
## "integer"
## Median.mortgage.repayment....monthly.
## "integer"
## State
## "character"
## X
## "integer"
## X.1
## "numeric"
## X.2
## "integer"
## X.3
## "numeric"2.2 Background and Assessment
The data came from a 2011 ABS (Australian Bureau of Statistics) dataset titled “Housing Arrangements - Homes Owned with a Mortgage in Local Government Areas”. Since the data is sourced from a Government source it is known that this is a very reliable data source for the public to use.
Being an independent Australian Government statutory agency, any political bias has been removed so that data doesn’t skew to support the party in governance(2011 being the time where Labor was in government) and it being an accredited source of data from the government, a significant portion of the data that they present to the public goes through various checks by various departments/people who fact check the datasheet various times before it is available for use from either Government ministers or to the general public. The only issue with the data set that may reduce it’s factual reliability is the fact that the data is quite outdated, as we are using 2011 data in the year 2020, and as we know housing in Australia in the past 9 years have changed dramatically, with it being a lot more expensive as time progressed. This means that the data in this dataset is not reliable with the current times and the current housing landscape.
The rows in this datasheet categorise rent and home ownership by different categories such as outright ownership or ownership with a mortgage, or rent through a real estate agent or a state/territory housing authority. Rows also contain the state of each Local Government Area(LGA) and median mortgage repayment by a monthly rate.
The columns in this datasheet cover the Local Government Area in each state (includes state that is in the LGA), as well as the values relating to the rent, ownership and median mortgage.
2.3 Stakeholders
In relation to the housing market in Australia, stakeholders of this data would include residents of Australia who are looking to either relocate or to buy their first home, as well as others which include real estate agents and real estate companies like LJ Hooker, McGrath and Century 21. Although the real estate industry has an interest in this housing data, residents of Australia who are looking for a relocation or to buy their first home will pay attention most as these will have a good indicator into which location is the best to move into, price-wise as information such as facilities nearby are not available in this data set.
2.4 Research Question 1: What are the trends regarding home ownership and home/apartment rentals in Australia by state?
For the purposes of this research question we created a new spreadsheet that was more simplified to cover trends in the real estate market more easily, trends such as total amount of people who rented a home/apartment, people who owned a home and other important factors come into play such as the average monthly mortgage in each state. The total amount of population in each state was also considered as an important factor to bring perspective into each state as we can make more implications from the data, such as total percentage of the population that owned/rented a house. Below are bar plots that show the data of the indicating factors that were mentioned here.
stateData = read.csv("C:/Users/georg/Documents/StateData.csv", header = TRUE)
barplot(stateData$Ownership, main = "Home Ownership in Australia", ylab = "Amount of Owners", xlab = "States", names = c("New South Wales", "Victoria", "Queensland", "South Australia", "Western Australia", "Tasmania", "Northern Territory"), las = 2)
barplot(stateData$Rent, main = "Renting in Australia", ylab = "Amount of Renters", xlab = "States", names = c("New South Wales", "Victoria", "Queensland", "South Australia", "Western Australia", "Tasmania", "Northern Territory"), las = 2)
barplot(stateData$Uncertain, main = "Unknown type of Home Residence in Australia", ylab = "Amount of Unknown Owners", xlab = "States", names = c("New South Wales", "Victoria", "Queensland", "South Australia", "Western Australia", "Tasmania", "Northern Territory"), las = 2)
barplot(stateData$Mortgage, main = "Average Monthly Mortgage in Australia", ylab = "Average Mortgage (In AUD/month)", xlab = "States", names = c("New South Wales", "Victoria", "Queensland", "South Australia", "Western Australia", "Tasmania", "Northern Territory"), las = 2)
barplot(stateData$Residents, main = "Total amount of Residents in Each State", ylab = "Population", xlab = "States", names = c("New South Wales", "Victoria", "Queensland", "South Australia", "Western Australia", "Tasmania", "Northern Territory"), las = 2)
Doing some calculations with the data given to us from these barplots, we can calculate the rent/ownership proportion with the total population in each state:
New South Wales: 67% own a property, 30% rent, 3% uncertain.
Victoria: 70% own a property, 27% rent, 3% uncertain.
Queensland: 63% own a property, 33% rent, 4% uncertain.
South Australia: 68% own a property, 28% rent, 4% uncertain.
Western Australia: 67% own a property, 29% rent, 4% uncertain.
Tasmania: 70% own a property, 26% rent, 4% uncertain.
Northern Territory: 48% own a property, 48% rent, 4% uncertain.
The implications we can make from the data are:
Victoria and Tasmania have both the highest proportions of home ownership at 70%.
The Northern Territory has both the lowest home ownership proportions and yet the highest rent proportions, both at 48%.
New South Wales, Victoria and Queensland lead every metric except for Queensland’s average monthly mortgage, with Queensland actually having the lowest monthly mortgage and Western Australia having the 3rd highest behind New South Wales and Victoria.
Savings.com.au, an Australian website that is designed to help Australians with their finances by giving information on general financial assistance, wrote an article in 2019 that used ABS data from July 2017 to June 2018. The article, written by Dominic Beattie, provided us with the updated home ownership and rental proportions, with the uncertain types of ownership excluded. The data received from the 2018 ABS data highlighted:
New South Wales: 64% own a property, 34% rent.
Victoria: 68% own a property, 29% rent.
Queensland: 63% own a property, 36% of people rent.
South Australia: 69% own a property, 30% rent.
Western Australia: 70% own a property, 28% rent.
Tasmania: 72% own a property, 26% rent.
Northern Territory: 59% own a property, 39% rent.
Comparing 2018 data to the 2011 data, whilst trends have remained mostly similar across the board, the Northern Territory’s property ownership has gone up, from 48% ownership to 59% ownership with rent going down from 48% to 39% and Tasmania now has the highest ownership percentage with 72% owning a property. The most likely reason behind this can be explained with the average housing cost explained in the article, which listed the Northern Territory and Tasmania having respectively the highest and lowest average housing cost, at $394 and $207.
2.5 Research Question 2: Is New South Wales’ average monthly mortgage payment higher than Australia’s total average mortgage payment?
This research question can be put into terms of an alternative hypothesis and a null hypothesis.
2.5.1 Hypothesis
\[H_{1}:\] The average monthly mortgage payments in New South Wales are higher than that of the average monthly mortgage payments in Australia overall i.e
\[M_{NSW} > M_{AUS}\]
\[H_{0}:\] The average monthly mortgage payments in New South Wales are either lower or equal to the average monthly mortgage payments in Australia overall i.e
\[M_{NSW} \leq M_{AUS}\]
To test this hypothesis, the alpha value 0.05 will be used to determine whether the alternative hypothesis or the null hypothesis is correct.
2.5.2 Assumptions
Some assumptions we can make from the data so far are:
We can assume that every instance of housing in Australia that has been observed so far is independent from each other.
Due to the fact that we are comparing one state of Australia with the entirety of Australia, we can assume that the two sample populations we are testing against do not have equal spread. To ensure we’re not testing two equal spreads of equal population, we can use a Welch-2 sample t-test.
2.5.3 Testing the hypotheses
First we calculate New South Wales’ average monthly mortgage. There are some difficulties with this due to the fact that the only available cost data is the median monthly mortgage, however it is still a decent indicator for the state’s cost of living. We find that in New South Wales that the average monthly mortgage in New South Wales is around 1603 AUD.
#All median mortgage numbers in every LGA in New South Wales.
mortgageNSW <- houses$Median.mortgage.repayment....monthly.[houses$State == "New South Wales"]
mortgageNSW## [1] 1450 1430 2167 2000 1733 867 2000 1625 1300 1300 1181 2100 1040 1419 1842
## [16] 1083 975 1000 2500 888 585 953 2167 1668 1398 2167 1800 2600 2000 867
## [31] 300 1517 1300 1300 1564 1018 1200 1300 780 1083 1300 1096 1083 1508 1637
## [46] 1430 1800 1127 932 1083 1517 1950 1517 1400 1200 1300 1517 1170 1300 1083
## [61] 888 1083 2034 1083 2058 2400 3000 2167 1200 955 1170 1300 2000 2167 3000
## [76] 1083 867 1733 2732 1300 3000 1495 1450 2167 1083 975 1733 3033 2477 1560
## [91] 1300 3033 1430 1040 1733 1208 1300 1062 1083 1777 2700 1517 1625 2162 1300
## [106] 2062 1985 3000 1650 1733 2000 2600 1300 2167 2317 1900 1517 2000 1517 2200
## [121] 2400 2537 1430 1083 1083 2578 1083 1300 1733 1600 1460 1235 500 1538 953
## [136] 867 810 867 2600 870 3000 867 1200 1200 2817 1868 2167 1950 3250 1733
## [151] 2000 1250mean(mortgageNSW)## [1] 1603.007We follow through the same process we did with New South Wales, however instead of one singular state we go through the whole country. We find that in Australia that the average monthly mortgage in all of Australia is around 1387.58 AUD. From this we can initially see that there’s a 215.42 AUD difference between New South Wales and all of Australia.
#All median mortgage numbers in every LGA in Australia.
mortgageAUS <- houses$Median.mortgage.repayment....monthly.
mortgageAUS## [1] 1450 1430 2167 2000 1733 867 2000 1625 1300 1300 1181 2100 1040 1419 1842
## [16] 1083 975 1000 2500 888 585 953 2167 1668 1398 2167 1800 2600 2000 867
## [31] 300 1517 1300 1300 1564 1018 1200 1300 780 1083 1300 1096 1083 1508 1637
## [46] 1430 1800 1127 932 1083 1517 1950 1517 1400 1200 1300 1517 1170 1300 1083
## [61] 888 1083 2034 1083 2058 2400 3000 2167 1200 955 1170 1300 2000 2167 3000
## [76] 1083 867 1733 2732 1300 3000 1495 1450 2167 1083 975 1733 3033 2477 1560
## [91] 1300 3033 1430 1040 1733 1208 1300 1062 1083 1777 2700 1517 1625 2162 1300
## [106] 2062 1985 3000 1650 1733 2000 2600 1300 2167 2317 1900 1517 2000 1517 2200
## [121] 2400 2537 1430 1083 1083 2578 1083 1300 1733 1600 1460 1235 500 1538 953
## [136] 867 810 867 2600 870 3000 867 1200 1200 2817 1868 2167 1950 3250 1733
## [151] 2000 1250 1300 1040 1300 1900 1300 1387 2422 1200 2400 1600 717 1300 1733
## [166] 1733 867 1252 1062 1900 1296 1647 975 2158 1146 1500 1300 1517 1517 1300
## [181] 1231 780 1800 1170 1712 1387 1950 1733 1200 739 1733 2000 1207 2000 1733
## [196] 2167 1733 1263 1600 1200 2000 2000 1517 1950 1733 1213 1300 1387 1950 910
## [211] 2200 999 1500 1300 1083 2474 1103 1800 1192 1083 1300 1495 1213 724 1960
## [226] 1875 1473 1800 2263 1647 650 0 1200 1625 867 453 953 500 2100 934
## [241] 1365 1300 884 1733 953 1315 2000 1350 0 1500 1047 844 780 303 650
## [256] 670 1387 2000 2080 1300 1343 1192 0 1733 1950 0 0 1517 1800 1465
## [271] 2163 975 0 1400 1928 0 2000 1083 0 1000 0 0 520 0 650
## [286] 2000 708 1690 1733 1500 1200 1300 1863 1426 1517 1733 0 1857 2000 1300
## [301] 1769 921 0 0 0 1863 1711 1366 0 1458 867 1083 2000 1600 1213
## [316] 1690 1200 746 563 1083 1044 1000 1120 1408 758 1300 1850 1083 802 575
## [331] 1062 1430 1200 1083 1374 0 1600 1083 1733 1607 1257 921 1200 1083 900
## [346] 1804 1473 867 563 1300 1599 1200 1305 1000 1777 1083 1404 2323 1408 780
## [361] 1116 1009 1517 874 1257 1950 1235 997 1950 975 1690 1300 758 1188 997
## [376] 1560 1850 954 1733 1842 1950 2000 1300 2043 1122 1500 1200 2600 900 451
## [391] 1647 1881 2600 1900 1993 700 1517 1408 2093 2708 2002 1586 1287 1524 693
## [406] 3000 800 1000 217 949 953 1500 1900 1359 1842 1413 700 781 660 2200
## [421] 364 1517 2167 2167 1647 1603 625 1000 1800 175 1842 1706 1000 2000 1976
## [436] 1950 1207 737 758 932 865 379 607 1800 900 953 800 1950 1300 442
## [451] 2167 2813 1140 332 900 900 2600 650 930 500 600 1842 6100 1950 1600
## [466] 650 1001 1200 3000 0 1300 1500 1350 2167 867 2500 1083 1300 2600 750
## [481] 1200 2000 3000 2000 2167 1005 2200 2000 2206 1950 272 532 1517 498 0
## [496] 2167 477 2300 927 2167 2104 1517 712 1439 611 1083 190 975 563 585
## [511] 2167 0 867 1404 953 1300 1148 1213 832 1097 1500 1200 1257 975 975
## [526] 1200 1083 1322 1700 1200 1127 867 1517 1300 1300 1300 1246 1275 1183 927
## [541] 1192 758 1300 1950 922 0 0 1300 2167 108 1733 2167 1625 2167 806
## [556] 542 834 2100 108mean(mortgageAUS)## [1] 1387.583testing <- t.test(mortgageNSW, mortgageAUS, var.equal = FALSE)
testing##
## Welch Two Sample t-test
##
## data: mortgageNSW and mortgageAUS
## t = 3.7905, df = 253.97, p-value = 0.0001879
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## 103.4998 327.3470
## sample estimates:
## mean of x mean of y
## 1603.007 1387.583We can see from the Welch two sample t-test that the p-value is less than that of the alpha value of 0.05. From here we can accept the alternative hypothesis, which is that the average of average monthly mortgage payments in New South Wales are higher than that of Australia’s average monthly mortgage payments.
3 References
Beattie, D.(2019, August 19). By the Numbers: Australian home ownership & tenancy. Retrieved November 12th, 2020. from savings.com.au ( https://www.savings.com.au/home-loans/by-the-numbers-australian-home-ownership-tenancy-statistics ).