Material Generation and Recycling Trends in the United States

Author

Rocío Rodriguez

Published

January 30, 2026

1.Introduction

This project asses historical information on materials in the U.S. municipal waste stream from 1960 to 2018 (measured in tons) from EPA. It includes data on the waste generation, recycling, and composting of the materials in question.

The main aim is to assess how materials waste generation has evolved in comparison with waste management practices.

The questions to be answered are :

  1. How has the generation of waste materials in the United States evolved between 1960 and 2018?
  2. Which materials have experienced the greatest increase in generation over this period?
  3. How does this increase compare with the corresponding recycling or composting rates?

2. Methods

2.1 Data Import and Cleaning

Code 
library(tidyverse)
library(here)
library(dplyr)
library(ggplot2)
library(gt)
library(gtsummary)
library(knitr)
library(DT)
library(kableExtra)

The dataset was tidied by renaming columns for clarity, converting specific values in Yardtfood_recy to numeric, and reshaping the data from wide to long format. This was done using pivot_longer(), which organized multiple related columns into two new ones—material and mat_status—with their values in a single column called qty_ton. Finally, it was saved the cleaned data as a CSV file.

Code 
raw_data <- read_csv(here::here("data/raw/mat_raw.csv"))


raw_data <- raw_data |> 
  rename(
    Paper_gen = Paper,
    Glass_gen = Glass,
    Metals_gen = Metals,
    Plastics_gen = Plastics,
    Food_gen = Food,
    Yardt_gen = `Yard trimmings`,
    Allothers_gen = `All other`,
    Paper_recy = Paper_recycled ,
    Glass_recy = Glass_recycled ,
    Metals_recy = Metals_recycled ,
    Plastics_recy = Plastics_recycled ,
    Yardtfood_recy = `Yard trimmings, Food waste_Composted`,
    Allothers_recy = `All others_Recycled`
  )

raw_data <- raw_data |> 
  mutate(Yardtfood_recy = ifelse(Yardtfood_recy == "Neg.", 0,
        as.numeric(Yardtfood_recy)))

processed_data <- raw_data  |> 
pivot_longer(
cols = Paper_gen:Allothers_recy,
    names_to = c("material", "mat_status"), 
    names_pattern = "(.*)_(.*)",
    values_to = "qty_ton")

write_csv(processed_data,
          here::here("data/processed/my-processed-data.csv"))

2.2 Data transformation and visualization

To answer the three research questions, the following methods were applied:

  • A linear regression was applied to total waste generation over the study period to quantify long-term trends. In addition, the Compound Annual Growth Rate (CAGR) was calculated to summarize the average annual growth in waste generation.

  • Waste generation quantities (in tons) by material were analyzed using descriptive statistics, including mean, median, minimum, and maximum values. Line plots were used to assess temporal trends in the quantity of each material. Furthermore, the CAGR (expressed as a percentage) was calculated for each material to identify differences in growth patterns among materials.

  • The dataset was then prepared to produce a combined visualization showing total waste generation, total waste recycled, and the recycling rate. A line plot was used to display these three indicators simultaneously for total waste. Additionally, a separate plot was created to illustrate differences in recycling rates among materials over time, enabling a comparative analysis of recycling performance by material.

3. Results

3.1 Generation of waste materials in U.S. between 1960 and 2018

It is observed that waste generation increased over time, with an average compound annual growth rate CAGR ( Figure 1 ) of approximately 2% between 1960 and 2018

Code 
total_gen_year <- processed_data |> 
  filter(
    mat_status == "gen") |> 
  group_by(Year) |> 
  summarise(
    total_waste_gen = sum(qty_ton)
  )
  
 total_recy_year <- processed_data |> 
  filter(
    mat_status == "recy") |> 
  group_by(Year) |> 
  summarise(
    total_waste_recy = sum(qty_ton)
  )
 
 #Calculus of CAGR for total waste generated
years <- 2018 - 1960
cagr_gene <- (total_gen_year$total_waste_gen[total_gen_year$Year == 2018] /
         total_gen_year$total_waste_gen[total_gen_year$Year == 1960])^(1/years) - 1
cagr_round <-  round(cagr_gene * 100, 2)
# Plot of CAGR
ggplot() +
  labs(title = "CAGR of total waste generated between 1960 adn 2018") +   xlim(0, 1) +  
  ylim(0, 1) +  
  theme_void() +  
  annotate("text", x = 0.5, y = 0.5, 
           label = paste("CAGR =", cagr_round, "%"), 
           size = 8, color = "black")
Figure 1: CAGR of total waste generated

Figure 2 illustrates that total waste generation follows a strong linear trend, with an 𝑅^2 coefficient of 0.97, indicating a highly consistent increase over time. Based on the regression results, an average annual increase of approximately 3,442 tons is observed.

Code 
# Fit the linear model
Waste_evol <- lm(total_waste_gen ~ Year, data = total_gen_year)

# Extract coefficients
coef <- summary(Waste_evol)$coefficients
intercept <- coef[1, 1]
slope <- coef[2, 1]
r_squared <- summary(Waste_evol)$r.squared

# Create the equation and R-squared string
equation <- paste("y =", round(intercept, 2), "+", round(slope, 2), "* Year")
r_squared_text <- paste("R² =", round(r_squared, 3))

# Create the ggplot
ggplot(total_gen_year, aes(x = Year, y = total_waste_gen)) +
  geom_point() +
  geom_smooth(method = "lm", se = TRUE) +
  labs(title = "Linear Regression: Total Waste Generation") +
  annotate("text", x = 1970, y = 250000, label = equation, size = 4, color = "black") +
  annotate("text", x = 1970, y = 190000, label = r_squared_text, size = 4, color = "black")
Figure 2: Linear regresion of total waste generation over time

3.2 Comparing waste generation by material

The results of CAGR show that plastic is the material with the largest increase over time, followed by the category “other materials,” which includes rubber, leather, textiles, and wood. Food waste ranks third in terms of long-term growth. This is summarize in the Table 1.

Code 
mat_gen_summa <- processed_data |> 
  filter(mat_status == "gen")

summary_materials <- mat_gen_summa |> 
  filter(Year %in% c(1960, 2018)) |> 
  group_by(material, Year) |> 
  summarise(quantity = sum(qty_ton, na.rm = TRUE), .groups = "drop") |> 
  tidyr::pivot_wider(
    names_from = Year,
    values_from = quantity,
    names_prefix = "q_"
  )
summary_materials <- summary_materials |> 
  mutate(
    CAGR_percent  = ((q_2018 / q_1960)^(1 / (2018 - 1960)) - 1)*100
  )

summary_materials |>
    gt() |>
    tab_header(title = "Compound Annual Growth Rate (CAGR) in percent for each material",
               subtitle = "q_year = Total waste in tons") |>
    fmt_number(columns = material:CAGR_percent, decimals = 0)
Table 1: Compound Annual Growth Rate (CAGR) in Percent for Each Material
Compound Annual Growth Rate (CAGR) in percent for each material
q_year = Total waste in tons
material q_1960 q_2018 CAGR_percent
Allothers 8,002 52,910 3
Food 12,200 63,130 3
Glass 6,721 12,250 1
Metals 10,818 25,600 1
Paper 29,985 67,390 1
Plastics 391 35,680 8
Yardt 20,000 35,400 1

Comparing waste generation by material in term of quantity not grow, paper is the most large waste in the US, follow by yardtriming, all others and food. This can be suported by mean and medians illustrated in Table 2 and Figure 3.

Code 
waste_tbl_gene <- mat_gen_summa |>
  group_by(material) |> 
    summarise(
        mean = mean(qty_ton),
        sd = sd(qty_ton),
        median = median(qty_ton),
        min = min(qty_ton),
        max = max(qty_ton)
    )

waste_tbl_gene |>
    gt() |>
    tab_header(title = "Total waste generation by material",
               subtitle = "Data between 1960-2018") |>
    fmt_number(columns = material:max, decimals = 0)
Table 2: Summary statistics for waste material generated
Total waste generation by material
Data between 1960-2018
material mean sd median min max
Allothers 27,671 14,468 27,911 8,002 53,090
Food 22,652 11,363 13,400 12,200 63,130
Glass 12,244 1,949 12,612 6,721 15,355
Metals 16,848 3,953 15,661 10,818 25,600
Paper 63,255 16,878 67,480 29,985 88,260
Plastics 16,171 12,065 15,871 391 35,680
Yardt 28,976 4,727 30,000 20,000 35,400
Code 
ggplot(mat_gen_summa, aes(x = material, y = qty_ton)) +
  geom_boxplot(fill = "lightblue", outlier.colour = "red", outlier.shape = 16, outlier.size = 2) +
  labs(title = "Box Plot of Waste generation by material", 
       x = "Material", 
       y = "Waste generated (tons)") +
  theme_minimal()
Figure 3: Box plot of total waste by material

3.3 Generation and Recycling of waste

By comparing waste generation and recycling through recycling rate percentages in Figure 4, it is observed that recycling levels remain substantially lower than waste generation in terms of total tons. This difference is illustrated by the recycling rate curve, which reaches a maximum value of approximately 35%. The results also show that the recycling rate increased from about 5% to 35% over a period of 58 years, corresponding to a compound annual growth rate (CAGR) of approximately 5%.

Code 
joined_totals <- right_join(total_gen_year, total_recy_year, by = "Year")
joined_totals <- joined_totals |> 
  mutate(
    recycling_rate = (total_waste_recy / total_waste_gen) *100
  )
scale_factor <- max(joined_totals$total_waste_gen, na.rm = TRUE) / 100

# Initialize the plot
ggplot(joined_totals, aes(x = Year)) + 
  
  geom_line(aes(y = total_waste_gen, color = "Waste Generated"), linewidth = 1) +
  
  geom_line(aes(y = total_waste_recy, color = "Waste Recycled"), linewidth = 1) +
  
  geom_line(
    aes(y = recycling_rate * scale_factor, color = "Recycling rate"),
    linewidth = 1,
    linetype = "dashed"
  ) +
  
  scale_y_continuous(
    name = "Total Waste (tons)",
    sec.axis = sec_axis(
      ~ . / scale_factor,
      name = "Recycling rate (%)"
    )
  ) +
  
  scale_color_manual(
    values = c(
      "Waste Generated" = "black",
      "Waste Recycled" = "blue",
      "Recycling rate" = "red"
    )
  ) +
  
  labs(
    title = "Waste Generation, Recycling, and Recycling Rate Over Time",
    x = "Year",
    color = ""
  ) +
  
  theme_minimal()
  
  #Calculating CAGR for recycling 
years <- 2018 - 1960
cagr_recy <- (total_recy_year$total_waste_recy[total_recy_year$Year == 2018] /
         total_recy_year$total_waste_recy[total_recy_year$Year == 1960])^(1/years) - 1
Figure 4: Comparing total Waste generated and recycled

By comparing the recycling rates by material in Figure 5 , it is observed that paper exhibits the largest and most consistent increase over time. However, for most of the other materials, an inflection point in recycling rates is evident between 1980 and 2000. In contrast, plastic consistently presents the lowest recycling rate throughout the entire study period.

Code 
# Preparation of data, combining food and yard trimming generation amounts to one new categoric
recy_rate_mat <- processed_data |> 
  mutate(material = ifelse(material %in% c("Food", "Yardt"), "Yardtfood", material)) |> 
  group_by(Year, material, mat_status) |>  
  summarise(qty_ton = sum(qty_ton, na.rm = TRUE), .groups = "drop") 

# Creating a new data frame with the new categoric Yardtfood to calculate recycled rates for each material 
recy_rate_mat <- recy_rate_mat |> 
  group_by(Year, material, mat_status) |> 
  summarise(qty_ton = sum(qty_ton, na.rm = TRUE), .groups = "drop") |> 
  pivot_wider(
    names_from = mat_status,
    values_from = qty_ton,
    values_fill = 0
  )
recy_rate_mat <-recy_rate_mat |> 
mutate(recy_percent = (recy/gen)*100)


# Filterng NA and INF values
recy_rate_plot <- recy_rate_mat |> 
  filter(!is.na(recy_percent) & is.finite(recy_percent))



# Ploting recycling rate by material over time 
ggplot(recy_rate_plot, aes(x = Year, y = recy_percent, color = material)) +
  geom_line(linewidth = 1) +
  labs(
    title = "Recycling Rate by Material Over Time",
    x = "Year",
    y = "Recycling rate (%)",
    color = "Material"
  ) +
  theme_minimal()
Figure 5: Comparing Recycling rates by material between 1960 and 2018

4. Conclusions

  1. Waste generation experienced a global annual growth, with a compound annual growth rate (CAGR) of 2%. Additionally, based on the regression results, an average annual increase of approximately 3,442 tons is observed.
  2. An inflection point in recycling rates was observed between 1980 and 2000, during which metals, glass, and organic waste experienced the largest increases. This change was likely associated with the introduction of new recycling technologies and the strengthening of municipal waste management systems during that period, as noted by (Ross and Law 2023a, 2023b).
  3. Plastic consistently exhibited the lowest recycling rate throughout the study period. In contrast, paper, the largest waste stream in terms of quantity, also achieved the highest recycling rates, a pattern explained by (Kinnaman 2000) due to well-established paper recycling programs and infrastructure.
  4. Although total waste generation grew at an average rate of 2% per year, recycling increased more rapidly, with a CAGR of 5% annually, indicating a relative improvement in the system’s capacity to recover materials over time. However, this trend applies to the global system overall and does not necessarily reflect improvements for all individual materials.

References

Kinnaman, Thomas C. 2000. “Explaining the Growth in Municipal Recycling Programs.” Public Works Management & Policy 5 (1): 37–51. https://doi.org/10.1177/1087724x0051004.
Ross, David E, and H James Law. 2023b. “Evolution of Municipal Solid Waste Management in North America.” Waste Management & Research: The Journal for a Sustainable Circular Economy 41 (7): 1177–87. https://doi.org/10.1177/0734242x231173308.
———. 2023a. “Evolution of Municipal Solid Waste Management in North America.” Waste Management & Research: The Journal for a Sustainable Circular Economy 41 (7): 1177–87. https://doi.org/10.1177/0734242x231173308.