[Tex/LaTex] Combining tabularx and longtable

tables

This is my first post so forgive me if I do something incorrectly.

I am trying to create a long table that goes over multiple pages but I don't want to lose the tabularx code that I've been using. In addition, I would like to shorten my first two columns (but still have them go over multiple rows if need be.

Any help would be appreciated and forgive me if this is a dumb question, I only started using Latex last week.

 \begin{table}[htbp]
    \centering
    \caption{Summary of Multipliers from the Literature}\label{Lit Summary table}
    \begin{tabularx}{\textwidth}{XXXX}
    \hline\hline
    \textbf{Study} & \textbf{Geographical Location and Level} & \textbf{Identification} & \textbf{Multiplier Result} \\ \hline \hline
        \citeA{nakamura2014fiscal} & US State Level & Instrumented on the fact that when national spending on the military rises by 1 percentage point of GDP in the US, different states, depending on their exposure to military spending, experience different levels of military build-up & 1.5 growing to 2.0 at zero lower bound \\ \hline
        \citeA{ramey2011identifying} & US Federal Level & Looked at the military build-up during significant wars throughout the 20\textsuperscript{th} century in the US &  0.6-1.2 \\ \hline \citeA{Barro} & US Federal Level & Spending multipliers are identified primarily from variations in defense spending, especially changes associated with buildups and aftermaths of wars. & 0.4-0.5 contemporaneously, 0.6-0.7 2 years later and if multiplier is permanent then this adds 0.1-0.2 to the multiplier  \\ \hline \citeA{auerbach} & US Federal Level & Extended SVAR approach & 0-0.5 in Expansion and 1.0-1.5 in Recession \\ \hline \citeA{fazzari2015state} & US State Level & Bayesian model comparison and generalized impulse response analysis to test for nonlinearities in the responses of output to government spending & 0.8 for states with low slack and 1.6 for states with high slack \\ \hline \citeA{Ramey2018} & US Federal Level & Analysed quarterly US data over a 120 year period using the local projection method developed by \citeA{Jorda} & 0.3-1.5 depending on the methodology and robustness check they used.  \\ \hline \citeA{shoag2010impact} & US State Level & Instrumented on the windfall gains that US states receive from investing public pension funds & Baseline multiplier over 2, rising to over 3 in times of economic slack \\ \hline \citeA{suarez2016estimating} & US County Level & Instrumented on Federal transfers to local counties due to changes in population forecasts between Census and Non-census years & 1.7-2 rising in counties experiencing economic slack \\ \hline \citeA{michaillat2014theory} & US Federal Level & Simple search-and-matching model to highlight the key economic forces of the multiplier & The multiplier doubles when unemployment rises from 5\% to 8\% \\ \hline \citeA{clemens2012} & US State Level & Instrumented on the variation in the strictness states' balanced budget requirements to determine the level as a means of measuring the level of spending cuts during a recession & 0.4 however it may be higher if states receive windfall funding \\ \hline \citeA{Ilzetzki} & 44 countries around the world \footnote{20 high-income and 44 low income} & SVAR approach on a novel dataset & High income countries tend to have a statistically significant positive fiscal multiplier while the opposite is true of developing countries and Investment in infrastructure could lead to a higher multiplier \\ \hline \citeA{acconcia2014mafia} & Italy Local Level & Looked at a situation when local governments were dismissed and public funding was severely reduced in Italy is response to being infiltrated by the mafia.   & 1.5 on impact, growing to 1.9 when dynamic effects are included. \\ \hline \citeA{fishback2015multiplier} & US State Level & Panel of annual Federal expenditure in each State during the 1930s & 0.96 when Federal transfer payments are excluded falling to 0.83 when they are included  \\ 
    \end{tabularx}
\end{table}

Best Answer

\documentclass[a4paper]{article}
\usepackage{geometry}
\usepackage{xltabular}
\def\citeA#1{}% For demo here
\begin{document}

\begin{xltabular}{\linewidth}{lp{3cm}XX}
\caption{Summary of Multipliers from the Literature}\label{Lit Summary table}\\ \hline\hline
\textbf{Study} & \textbf{Geographical Location and Level} & \textbf{Identification} & 
\textbf{Multiplier Result} \\ \hline \hline
\endfirsthead
\hline\hline
\textbf{Study} & \textbf{Geographical Location and Level} & \textbf{Identification} & 
\textbf{Multiplier Result} \\ \hline \hline
\endhead
\hline
\endfoot
\citeA{nakamura2014fiscal} & US State Level & Instrumented on the fact that when national spending on 
        the military rises by 1 percentage point of GDP in the US, different states, depending on their 
        exposure to military spending, experience different levels of military build-up & 1.5 growing to 2.0 
        at zero lower bound \\ \hline
        \citeA{ramey2011identifying} & US Federal Level & Looked at the military build-up during significant 
        wars throughout the 20\textsuperscript{th} century in the US &  0.6-1.2 \\ \hline \citeA{Barro} & US 
        Federal Level & Spending multipliers are identified primarily from variations in defense spending, 
        especially changes associated with buildups and aftermaths of wars. & 0.4-0.5 contemporaneously, 
        0.6-0.7 2 years later and if multiplier is permanent then this adds 0.1-0.2 to the multiplier  \\ 
        \hline \citeA{auerbach} & US Federal Level & Extended SVAR approach & 0-0.5 in Expansion and 1.0-1.5 
        in Recession \\ \hline \citeA{fazzari2015state} & US State Level & Bayesian model comparison and 
        generalized impulse response analysis to test for nonlinearities in the responses of output to 
        government spending & 0.8 for states with low slack and 1.6 for states with high slack \\ \hline 
        \citeA{Ramey2018} & US Federal Level & Analysed quarterly US data over a 120 year period using the 
        local projection method developed by \citeA{Jorda} & 0.3-1.5 depending on the methodology and 
        robustness check they used.  \\ \hline \citeA{shoag2010impact} & US State Level & Instrumented on the 
        windfall gains that US states receive from investing public pension funds & Baseline multiplier over 
        2, rising to over 3 in times of economic slack \\ \hline \citeA{suarez2016estimating} & US County 
        Level & Instrumented on Federal transfers to local counties due to changes in population forecasts 
        between Census and Non-census years & 1.7-2 rising in counties experiencing economic slack \\ \hline 
        \citeA{michaillat2014theory} & US Federal Level & Simple search-and-matching model to highlight the 
        key economic forces of the multiplier & The multiplier doubles when unemployment rises from 5\% to 
        8\% \\ \hline \citeA{clemens2012} & US State Level & Instrumented on the variation in the strictness 
        states' balanced budget requirements to determine the level as a means of measuring the level of 
        spending cuts during a recession & 0.4 however it may be higher if states receive windfall funding \\ 
        \hline \citeA{Ilzetzki} & 44 countries around the world \footnote{20 high-income and 44 low income} & 
        SVAR approach on a novel dataset & High income countries tend to have a statistically significant 
        positive fiscal multiplier while the opposite is true of developing countries and Investment in 
        infrastructure could lead to a higher multiplier \\ \hline \citeA{acconcia2014mafia} & Italy Local 
        Level & Looked at a situation when local governments were dismissed and public funding was severely 
        reduced in Italy is response to being infiltrated by the mafia.   & 1.5 on impact, growing to 1.9 
        when dynamic effects are included. \\ \hline \citeA{fishback2015multiplier} & US State Level & Panel 
        of annual Federal expenditure in each State during the 1930s & 0.96 when Federal transfer payments 
        are excluded falling to 0.83 when they are included  \\ 
\end{xltabular}

\end{document}